Episode 117: PASS Summit Retrospective

Have you ever had a week without a Wednesday?  Me neither, so for this episode, we recap our experience at PASS Summit and talk about all the amazing compañeros we met and give some thoughts around our very first meetup.  It was so much fun to meet everyone and both Steve and I gave sessions, which was a great experience–and a first for me.

One, we both realized after the fact is, we are not good at social media or taking pictures!  We apologize we don’t have a bit more to share on that front, but hey–we do have faces for radio.

Episode Quotes

“I find when I’m at a conference I get so busy and I enjoy talking with people. I just don’t think to take out the cameras down for shot.”

“It’s nice to have live music but at the same time it made a little difficult to hear sometimes.”

“Watching these keynotes from the hotel is super convenient.”

“Good questions that’s always make you feel good as a speaker.”

“Be prepared, bring two laptops and a backup plan, and a backup to your backup plan.”

“The last 5 minutes before a session starts, even if everything is ready to go and going smoothly, the last 5 minutes is the stress factor.”

 

Listen to Learn

02:32 Companero Shoutouts

07:47 SQL Server in the News

10:50 PASS Summit Retrospective, being on social media

12:47 Tuesday night at Seattle – Carlos’ Mardi Gras suit, the live music

15:56 Volunteer Party18:46 Thoughts about the livestreaming of keynotes

20:13 Wednesday sessions, SQL Trail Mix event

25:10 Thursday – Steve’s session on Database Corruption

31:15 Thoughts about the Game Night

32:29 The Sponsor Area

34:07 Friday – Steve’s session (Backup and Corruption) and Carlos’ session (Establishing Baselines)

40:23 End of conference and unfortunate story at dinner

Transcription: The PASS Summit Retrospective

Carlos: Companeros! It’s nice to have you on the SQL Trail again.

Steve: Yes, and welcome to Episode 117. This is our PASS Summit retrospective episode, which I’ll tell you we got to meet a lot of companeros last week.

Carlos: Yes, and it was great meeting a lot of you and getting to chat with you. I feel like I’ve been doing a little bit of retrospective, a couple of retrospectives here. It is the conference season so we’re kind of giving our thoughts around the conference, what went well and what we enjoyed being out there in Seattle.

Steve: And I know I say last week, but yeah, it was last week when we’re recording this but it will be two weeks back when it actually airs.

Carlos: That’s right. Man, how time flies.

Steve: Yeah, and I guess with you being on the East Coast and me being on the West Coast we don’t have that many opportunities where we’re actually physically in the same location.

Carlos: That’s right. And we did, again you’ll see as we kind of get into this just some of the activities that we had. But our schedule was jam-packed and some client issues kind of rose and we didn’t have as much time as I was hoping. So for example we didn’t get to do an episode, I don’t want to say live, but face to face, which has been something that we thought about but didn’t plan well. But this works and hopefully the companeros will enjoy it.

Steve: Maybe next time we’ll have to have you come a day earlier or something just to do that. Or I’ll go a day early depending on where we’re going to.

Carlos: Right, there you go. There is always next year, right, sound like a bad sports team. Ok, so getting into some companero shoutouts, and because of Summit and running into so many of you. We’re actually going to break this up a little bit, so if we don’t mention your name now don’t freak out just yet because we may be working some of that into the actual story or our retrospective. So first we have Justin Wailey, who I think I incorrectly identified as being from Georgia. His Twitter account says he is in Alabama. It was good to connect with Justin. I had met him two years ago. I think it was his first Summit back then and it was good to reconnect with him.

Steve: Yup, and we’ve got Tom Norman on the list too. I don’t remember meeting Tom, do you?

Carlos: Yes. He was on our Trail Mix, but I did run into him at the opening social. He was a buddy and so we chatted. Tom is here from the East Coast and it was nice to connect with him again. Now another one that we both talked to kind of in sessions so actually in the convention center, and it’s Brent Unruh. I’m pretty sure I’m pronouncing that right. He told me how to pronounce it. Hopefully I’m getting that right Brent. But BrentU on Twitter and he is the one who brought us the question because many people talked about flying and his question was, “Would you still choose flying if you can only go 3 miles an hour?”

Steve: Oh yeah, that was long time back on the podcast, SQL Family session wasn’t it?

Carlos: That’s right. So you’re long time listener Brent. Thanks for reaching out and connecting with us.

Steve: Yup, and then we also talked with Andy Levy from Rochester.

Carlos: Yes, and we’ll probably bring him back up maybe when we get to the sessions but nice to meet him. Contributor on the DBA Tools team and it was nice to connect with him.

Steve: Yup, and I know he and I have been back and forth on LinkedIn a little bit since the conference talking about Database Health Monitor.

Carlos: Oh, got you. Ok. Another DBA Tools regular is Constantine Kokkinos. It’ was good to meet him. And he is the one who got me the, well I guess so Andy had them. He was the one who brought them but I love the DBA Tools.

Steve: You know, there were quite a few members of the DBA Tools group there and it was good to meet Constantine.

Carlos: Yes. I don’t think Claudio was there but I did talked with Rob. Unfortunately I did not meet Chrissy, wanted to do that but for a reason didn’t happen.

Steve: Oh! Yeah, I got to meet Chrissy in the speaker room. Sort of they are practicing. I heard her talking and I did not know what she looked like but I just totally recognized her voice from the podcast.

Carlos: Oh, got you. Ok.

Steve: So I got to say hi and met her there as well.

Carlos: Oh very nice. And then a couple of our MVPs want to give shoutouts to Argenis Fernandez, Brent Ozar, Cathy Kellenberger, so thanks for coming out to the Trail Mix event and talking with us, giving some of your thoughts. I know everyone appreciate getting your thoughts on some things and mingling with us so we do appreciate that.

Steve: Yeah, it was an interesting group to get to know there.

Carlos: Sean Barry is another one. Good to meet him. Kevin Wilkes, long time supporter and previous guest on the show. Adam Jacobson who is actually someone who have introduced us to a client and happen to be there, so it was nice to be able to catch up with him.

Steve: Yup, I did not get to talk to him as much as I would like to but we did talk briefly.

Carlos: Yes, and that was a reoccurring theme I think for Summit is, and that was a surprise. I mean, particularly with the podcast and that’s really where and at the SQL Saturdays I feel I want to be connected. That’s one of the things that I want people to know me for if you will is being connected. And I was surprised to just how many people I wasn’t able to talked to, so I got like empathic, Oh yeah, I got to follow up with them and ended up, I’m at home.

Steve: Well, you know, on that point. I mean, you talk to some people and they think, “Oh, you’re going to a conference for the week. It’s like a vacation. Time to relax.” But, holy cow, it was busy. I mean, one, it never stopped it seem like which is awesome. I mean, I really enjoyed it but it’s certainly no vacation. It’s busy.

Carlos: That’s right, that’s right, and then from our perspective as well we are trying to, like everyone. Not saying that it’s unique to us but trying to understand the ramifications of all the new features that Microsoft is dumping out there on us, the keynotes. We had that keynote on the Cosmos DB. That’s something that I think all of us are trying to understand, ok, what’s our future is going to be like. What’s all looking like and understanding some of those concepts. That takes brain cycles as well.

Steve: And speaking of that future, I mean does that bring us right into SQL Server News?

Carlos: It does. Yes, let’s do the SQL Server in the News. So we ran into Travis, and Travis he’s the PM for the Linux migration and I’m not exactly sure how his role has evolve. I think it’s still something that he is still doing. But we were chatting and of course they had mentioned many of the new features. We won’t cover all of them today. We’ll save some of those and kind of spread them out for future episodes. What he mentioned to me and he said that they may be announcing everything that they have just put out and SQL Server obviously became live at the beginning of the month. They are going to announce VNext as potentially as early as next week.

Steve: Wow! So if you’re not on SQL Server 2017 yet you will soon be behind.

Carlos: Yeah, that’s right. You just made to 2018.

Steve: VNext will be taking over in the near future perhaps 2018.

Carlos: Yeah, so again, just kind of continuing that cycle. And you used to think about the platform going on to Linux and all the additional features that they are trying to integrate there and what the monumental challenge that is.

Steve: Oh yeah. It’s really interesting because part of it is there are so many new things going into SQL Server at the rate that it’s going out right now and other Azure related databases and things. I mean, it’s really tough to figure out, ok what are the things that are really going to stick and are going to be the fundamental pieces that we’re going to need to know going forward, and what are the things that are maybe a little bit of hype and maybe don’t end up being that big of a deal.

Carlos: You know that’s right. I think the comment is we don’t need to feel. And even us as consultants I feel like we’re in a pretty good position because we get to be lots of different environments and touch of this new technology sometimes. But a lot of times I can feel like I’m in the Stone Age because all the new things that people are trying to do. You know, cloud this and cloud that. I think companeros you don’t need to be afraid if you’re not on VNext. Don’t despair.

Steve: Right. And I think, but that’s part of the benefit of the conference too is you go and you’re exposed to a lot of this new information. And it sort of pushes you out of that comfort zone that you’ve been in for the last couple of years perhaps.

Carlos: Sure, exactly. And if nothing more like to be educated how things are changing and the types of problems that people are trying to solve. Your current organization may be not trying to solve that problem but if you find a problem that you’re then interested in again you’re creating value and then trying to go out and solve that for a different place or a different organization. Who knows what will happen, but being in a position to be able to do that is valuable.

Steve: Yeah, absolutely.

Carlos: Ok, so ultimately today it is Steve and I are doing the retrospective, we will have some shownotes. Now, I would say pictures and I guess we can get into this a little bit but we as busy as we were you could tell that we’re not great social media people, and do not take nearly enough pictures.

Steve: Yup, and you know, that’s the thing I find when I’m at a conference I get so busy and I enjoy talking with people. I just don’t think to take out the cameras down for shot where like there are others who say, “Oh look, I got a picture with this person with that person.” Wow, I just didn’t have time to do that. I think I need to make time for that next time.

Carlos: I think we are showing our age there a little bit, right? We’re not the selfie first kind of people.

Steve: Yeah.

Carlos: And that’s like for Twitter. I mean, I like to be on Twitter and put some things out there but I’m definitely not one of those people that has to put my every waking moment detail on there. And then you get busy if that’s not the habit, right? You haven’t trained yourself to do that then the day goes by and you’re like, “Opps.” In fact today I was actually replying to some tweets that happened during the Summit, so that was kind of embarrassing.

Steve: Yeah, it’s like you need an extra 4 hours of each night to go back and catch up on everything.

Carlos: That’s right. My old man disease is already kicking in pretty good. Particularly in the West Coast time, right, I’m there in Seattle or being that my body is on East Coast time rather. I’m in Seattle, I’m ready to go to bed. But the shownotes episodes are going to be at sqldatapartners.com/summit.

Steve: Or at sqldatapartners.com/117 for the episode number.

Carlos: Yeah, so as far as some of the other things that we did. So Tuesday was kind low key, actually we’re in Seattle. I know that there was like the user group meeting or the chatgroup leader meetings or whatnot. We actually had some client engagement meetings so we took advantage some of that time particularly with me being on the East Coast to do a little bit of that. But we didn’t get to check in and I started meeting people.

Steve: Yup, and on Tuesday night there was the Welcome Reception for everyone, and we went to that. That was a lot of fun and got to see a number of people that I hadn’t seen in a year or so.

Carlos: Yes, and if you missed me, I’m sorry I was trying to stand out as much as possible in the Mardi Gras suit.

Steve: Right, some people might have thought you were an entertainer of sometime with that suit Carlos rather than actually an attendee.

Carlos: Yes. I was accused more than ones, “Are you hired to…” “Nope. Just me.”

Steve: Well, that was a pretty awesome suit and if you haven’t seen Carlos in his I guess Mardi Gras suit we could call it?

Carlos: Yes, that’s right.

Steve: Yeah, take a look online and there are some pictures of him on that.

Carlos: Yes, we actually get those up on social media. Now, one thing I thought was interesting and I don’t remember. Now, I missed last year’s Summit. I don’t remember them doing live music at the welcome.

Steve: No, that was different because last year I think there was not a live band last year. It was recorded music and they were doing like slideshow or something like that I think last year.

Carlos: So I thought it’s nice to have live music but at the same time it made a little difficult to hear sometimes.

Steve: Right, and we are standing chatting with someone and you realized that it’s so loud you’ve got to be six inches from their face just to hear them. It makes it tough to communicate. If I had any say in it, I would love to say, maybe we’re getting old, but turn down the music. That sounds like an old man story there.

Carlos: Yeah, I really have to rename this “The Old Man Review”.

Steve: I guess we’re showing through.

Carlos: I agree, it’s one of those things. I don’t know if we could have a little bit more time. Of course people come in and out of that as well, but maybe pausing that music or giving us an hour before it plays. I don’t think they did take a rest, but yeah, I don’t know.

Steve: And of course, Tuesday night was the Volunteer Party.

Carlos: Yes, and so again more old man. I stand at the welcome reception a little longer than you did. I know you and Derek went over to that. I did not want to walk over there. I guess I was too cheap to take an Uber. Oh and that was when Andy Leonard actually gave me a book. So I had this book that I’m like, “What the heck I’m going to do with that?” So I ended up taking it back to the hotel room to drop it off and then deciding, “No, I’m not going to make it out there.”

Steve: Yup. Well, so I ended up going to the volunteer reception or volunteer party I guess. What I like about it is a couple of things. One, free food. So getting to meet some of the other speakers, some of the other PASS chapter leaders, some of the other volunteers, I mean, it’s a great group of people. What I would say is sign up and volunteer in some way if you’re not part of that because then you could be part of it.

Carlos: Yeah, that’s right. And you meet the Costa Rican guys, right? The Hispanic guys?

Steve: Yeah. I guess there were 4 or 5 of them and I think I ran into them that same party last year. Yeah, good group of guys. Bright guys from Costa Rica.

Carlos: Yeah, good times. And I did the mistake of then trying to go out afterwards because the DBA Tools Team had like a reception I saw on Twitter. I’m like, “Oh, that’s just right here. I’ll go the block.” When I got there they were actually all headed to SQL Karaoke which I then felt stupid because I just got in there and it was ending. Here I didn’t want to walk to the volunteer thing so then I ended up walking all the way to SQL Karaoke and if you know where that is. It’s like on the borders of Seattle, so I walked a lot more than that what I wanted to that evening. But it was all good, so one of the guys I got to meet in line actually. Again, feeling stupid again because I didn’t hit to that ticket to get into SQL Karaoke so there was that whole thing. But the SQLBits, no not SQLBits. SQLBits is Rodney from Florida. They guy who wrote the SQL Server 2016 song.

Steve: Oh yeah. I forgot his name but he was at SQL Saturday at Redmond last year.

Carlos: Right, so he was in line there behind me and so we started chatting and that was a lot of fun.

Steve: And he was going to Karaoke even though he is a real musician.

Carlos: That’s right. But we ended up, “Ok let me see. I’ll show you how this is done.”
I also ran into, so in the opening reception another name, Bert Wagner. It was funny because I walked up to Bert and like. I know you, I can’t remember where I know you but I feel like we’ve met. And then it wasn’t until Wednesday he actually came to our SQL Trail event and like, “Ah! Now I remember from the sign up list.”

Steve: Oh yeah. And I remember meeting Bert as well and talking a little bit about his job and what he does. Yeah, definitely cool.

Carlos: Right. So one of the takeaways and one of the things now, I’m not sure that we’re maybe promoting this idea but we found that watching these keynotes from the hotel is super convenient. Now, that does mean that you’re kind of miss out on breakfast where you have to kind of go on your own way for breakfast. But that’s something that we did and just worked out really well.

Steve: Oh yeah. And what I liked about that was because normally you go to the keynote and it’s jam-packed full, and it’s hard to see. I mean, depending on where you’re sitting in it could be a challenge. So just watching from the live broadcast and watching from the hotel room then you’ve got the chance while you’re doing that when they mention something then go look it up. Go online and take a look at it while you’re watching it. And I think that I like doing it that way rather than being there where you’re just sort of stuck in a seat and glued to the room.

Carlos: Right. I’m not sure if it was my tablet or what. mood of your laptop. It seem to be fine but you do sometimes hit that streaming issues. But once they got it going, PASS did a good job on their side to at least make sure that plenty of streams were available.

Steve: Yup. So then Wednesday there were some great sessions and I know I got to see a couple of those. Then later in the day on Wednesday we had the SQL Trail Mix event.

Carlos: Yes, we weren’t quite sure what to make of this, right? How this is going to turn out and special thanks here to Sean and Jen McAllen. We actually reached out to them to get their thoughts they’ve been doing several events there at Summit. And we thought, ok, because you do hear about all these different events would there be room enough for ours.

Steve: Would people want to go to ours?

Carlos: Would people come? That’s right. So it’s Steve and I sitting there eating peanuts, so these are all kind of thoughts kind of growing through our head. So we said, “Ok, let’s do it. And we’ll reach out obviously to the companeros and to our podcast guests and we’ll kind of go from there. I think we were pleased with the turnout. Who was there, thank you, so if you were there companeros it was great to meet you. I guess there are a couple that we want to specifically point out here but we had a great time chatting with you.

Steve: And I think what really surprised me was the small amount of promotion that we did, the small amount of time we had to promote it, and how many people actually showed up in that small amount of promotion time.

Carlos: Right, so that is true. If you didn’t get an invite for whatever reason, so we apologize because we did fill up way faster than we thought it would. So again, we were kind of thinking 25 people would be super successful. That’s was kind of our number. We knew some people wouldn’t come. We ended cutting off registration at 45 and that happened in a matter of about three hours, so it was super quick.

Steve: Yup, and I guess just a little bit of background for those who don’t know that the SQL Trail Mix was. This was an event that we hosted where it was a restaurant/bar where we reserved some space. We got a number of tables and we bought appetizers for everyone although we didn’t buy drinks for everyone. Everyone was on the hook for their own drinks. And it was just show up and chat. Show up and meet other podcast listeners. Meet other podcast guests and give us feedback on the podcast. I had a great time. It was awesome.

Carlos: I agree. It was good. For example, rubbing shoulders with James, so James Youkhanis He has been on Twitter. We’ve exchanged emails. He has suggested several podcasts topics and so to finally been able to meet him, rub shoulders with him was actually pretty cool. The guys from Connecticut, Will Conklin and Chris Albert, long time listeners of the podcast, so we shoutout to you guys. Thanks for coming up and chatting with us. And then Josh Simar. I can’t remember all of a sudden where Josh is from, but he had mentioned that he’s been listening to the podcast and it was great to chat with him as well.

Steve: And then also, Andre Ranieri is a friend of mine that I’ve known through PASS events for quite a while and he showed up at the event and have a chance to catch him up a little bit.

Carlos: Right, and that was an interesting thing. While we had our registration, I don’t know, we probably ended up with 30-35 people, because people did come in and out. Registration kind of help but it didn’t indicate everyone so I guess there is some thought we need to put into how to do that. We just want to make sure that we could provide for everyone who did show up.

Steve: Yup, and I think if we’re going to do it again or I hope we do it again. I think maybe going a little bit bigger but not massive would be good.

Carlos: What was interesting because I think kind of worked to our favor potentially is that we didn’t realized that we’d also be going up against seven world series game. I think that impacted just the bar itself because when we talk, “Yeah, out Wednesdays are dead. Why don’t you come it would be great to have a group.” And then we get in there and I’m like this is dead.

Steve: Oh yeah, it was packed. It really was.

Carlos: Ultimately we would very much like to continue to meet up with you guys and I’m not sure how well this would work at other events but if we have opportunities to something like this of course we’ll try to do it again.

Steve: So then that brings us to Thursday.

Carlos: Yes, that’s when Summit got real in the sense because you had your session. You had the first of two sessions.

Steve: First of two sessions. This was the first time at PASS Summit that I actually had two sessions to present. The first one was on Database Corruption, and I was on early afternoon I think. It was one of those things where as a speaker you always worry about getting connected, at least I do. Is the presentation going to work? Is the laptop going to work and all that? The laptop I’ve had for about 2½ I’ve never had a problem. Everywhere I’ve gone, you plug it in, it work just fine. It’s even work fine at two former PASS Summits on similar equipment. But still every time we go I tried out. There is this speaker ready room that they have to try it out on their projector and make sure it works. Try that all out ahead of time and get there and it will not for some reason the multi-video output will not connect to the projector. And of course, I’m the kind of guy when I present, that as soon as the previous speaker is off the stage I am there taking every minute I can to make sure that it’s going to work. And it didn’t work, and it didn’t work.

Carlos: And the AV people came up and they were trying to help you.

Steve: Yup and some of our friends Derek Bovenkamp and Randolph West. I mean they were in the audience jumping in trying to help with, “Try this, try that. Here is the USB stick, things like that.” And finally we are about 7 or 8 minutes late when I started my session which I apologize for that and I felt so bad for doing that. But I finally just put it on a USB stick, put it on a different laptop and did the presentation from PowerPoint on the other laptop. Not optimal but what I learned on that I guess is be prepared. Back to the Boy Scout motto, right?

Carlos: We’re going to the Boy Scout motto, there you go.

Steve: Be prepared, which I thought I was but be prepared with two laptops, which isn’t a financial thing because I had an extra laptop sitting at home. I could have prepared and brought up with me. But be prepared, bring two laptops and a backup plan, and a backup to your backup plan because whatever you think might go wrong, whatever could go wrong would go wrong eventually and it did in that session. I think I got a lot of feedback at the end that people were happy with it but I just felt bad getting started 7 or 8 minutes late because of that technical difficulty.

Carlos: And thanks to PASS for providing another laptop, right, PowerPoint loaded that you can move over pretty quickly. I think it worked well. The way that you gave that was kind of a story based idea, concepts. It was an introductory session, and so as you told the stories that kind of walkthrough some of those things. I thought your audience was super interactive. They were engaged, they were, “Urgghh”. When you tell a struggle they made it their own struggle or you could sense that they have been there, those kinds of things. And so I thought that it went really well.

Steve: Yeah, and what was different with that session was an introduction to database corruption. I hadn’t plan to do that sort of deep dive type bit level corruption fixing that I’ve done in the past. If I had planned that it wouldn’t have gone so well with the technical difficulties we had. But because it was more sort of the high level what is corruption and how do you deal with it. I think it worked out there really well and it was different than how I would have presented corruption in the past as well. I had a good time with it and I think people enjoyed it. At the end of course, though everyone comes up to ask questions to the speaker and you’re trying to get off the stage and all that because there was another speaker coming in we took the conversation outside to the outside of the room and continued for about 20 minutes with different people that had questions about corruption. It was really cool. I had a good time with that.

Carlos: I think that’s when we ran into Brent as well, Brent Unruh. Yeah, so good questions that’s always make you feel good as a speaker. People would come up afterwards and want to continue the conversation. It’s a mix bag, right? As an attendee you want to get that conversation in, then there’s other people in line and you want to go off to your next session as well. And then as a speaker you want to try to accommodate as many people as you can but we know that some of you probably slip out without having to have that conversation, so we apologize. I think another reason why we want to have an event like that is so that we can have that connection time in a less time constrained environment.

Steve: Right. And also I’d like to give a special thanks on that to Randolph West because last year at my corruption presentation and this year with the corruption presentation, at the end when you’re done speaking and everyone comes up to ask questions he helps filled some of those questions. And that was really awesome because he gave me a chance to get a drink of water and listen to what are some of the other questions were while I was packing up my laptop at the end. Thanks Randolph, I appreciate that.

Carlos: Yes, very cool. So Thursday night, a couple of shoutouts to the guys that helped organize the game night. Now ultimately it’s done through PASS but Kevin Hill was instrumental. I guess that’s how I found out about it. No, that’s not true. I found out about it through you, and then after investigating it Kevin Hill was kind of helping promote it and organize it a bit. And then he and Matt Cushing, one of those I have interacted with on Twitter. It was nice to meet him and shake hands with him.

Steve: Yeah, the game night was interesting because you come in and it’s just a bunch of tables and board games. And you first walk in the room and it’s a bit intimidating to look at all the people playing board games and wonder, “Ok, where I’m going to sit down?” And you just kind of wander just around a little bit, “Hey, there is a game that’s starting right now. I’ll jump in and play.” And it was awesome. I had a good time.

Carlos: Right. I think that worked really well. I think there was some confusion because people didn’t realize they had to pay to go. And I think that $12 was to buy a drink. Again, as somebody who doesn’t drink alcohol maybe they could have done without that, but I think it’s one of those things people liked it. But it did cause some confusion for folks.

Steve: Right, and I think also some of the $12 of that Game Night goes to the cost of the staff coming in and taking down regular seating and putting up tables because that was in the presentation rooms, so they had to come in and change out the room for that event.

Carlos: Right, and that makes sense, so the venue probably have cost associated with that as well. You’re exactly right. That makes a lot of sense. And then all of a sudden we’re at Friday. Well, the other thing I guess I do want to say about Thursday before we get to Friday is that I didn’t spend a ton of time in the Sponsor Area but Thursday was the day that I was able to do a bit of that. I guess special shoutout to the Windox guys, so Paul and his team. Got to meet with his partner and talk with him a little bit. It seemed a little bit smaller this year that in years prior, the Sponsor Area.

Steve: Yeah. That was interesting and I wonder why. I think the big players were there that you normally see and they had big booths. And then some of the smaller players that you see like Sean and Jen from MinionWare. I mean, they had their booth there. But it sort of the overall it just didn’t seem quite as large as what I remember in the past. Maybe it was just a different layout that made it appear that way too.

Carlos: Yeah, perhaps. On the other hot side is that we’ve also gone to some very large conferences like the Microsoft Partner Conference or Ignite. These are like 25,000 people conferences, and so all of a sudden you get that magnitude, right? And not to say that, I can’t remember what the attendance was but between 4,000 and 5,000. Not to say that that’s small but you’re talking about a quarter of a size and so it can pack things pretty quickly.

Steve: And then Friday, so we both had sessions. I had a session that was right around like 11:00 – 12:30 so right before lunch, and that was on Backup and Corruption. And although my laptop didn’t plugged in quite as easily as I would hoped because I spend hours working on it the night before to make sure it was going to work. It worked, it was just something to do with the video switching, Switching from full screen PowerPoint over to Management Studio. I had a little difficulty with there but it was better than the day before, and the topic was on Backup and Recovery. The thing that I had a lot of fun with was automatic restore script where when you write your backups, it would write the restore script at the same time. So that if you ever have to recover from backup you just find the restore script associated with the backup or transaction log file, open it and then uncomment a couple of lines and you’re running your restore. That was sort of the crowd pleaser I think in that presentation and it was a lot of fun.

Carlos: Oh very nice. Yes, so I was running a bit behind and I know that I had offered my laptop as a substitute if you needed it and by the time I got in there, you’re on the agenda, the presentation was fixed, and your corruption session was full. But the backup session was even fuller. I mean, there were lots of people standing and I thought, “Hmm. I’m not going to be able to get in this one.”

Steve: No, no. So the laptop problem was resolved and the presentation went smoothly from that perspective so that was a lot of fun. So then your session, Carlos.

Carlos: Yeah, so I was at the end and.

Steve: Establishing Baselines, right?

Carlos: Yup, through monitoring or managing performance through baselines was that kind of idea but taking baselines and thoughts around that idea, and so the last session of the day. I mean, obviously at that point and people are Summit is now wearing on them and everybody is a bit tired. So first I will say thanks if you were there, so again, to Will and Chris. I know who were there. Of course you were there, Andy was there, Mariano came, so thanks for you for coming out. That was my first time speaking at Summit and I feel like I do pretty well in group presentations. I’ve given lots of them and feel at home. I wasn’t so nervous and so I actually got up on the platform. And for whatever reason, that visual like being elevated. It was like, “Whoah.”

Steve: I’ll tell you. I personally think the last 5 minutes before a session starts, even if everything is ready to go and going smoothly, the last 5 minutes is the stress factor, right there. And then once you start talking. I mean I could see you getting nervous ahead of time, once you start talking then it really smooth over at that point and everything was cool I think.

Carlos: Right, so to help us some of that. Like I tried, which didn’t seem to go pretty well was creating a PowerPoint or some way to potentially engage the audience a little bit beforehand. Now I started just asking some questions so I had a little PowerPoint. I call it “Where in the world?” just like a photo and then the idea was you guess where it was. That didn’t seem to engage to many people which I thought was curious.

Steve: But there were a lot of interesting pictures that kept people occupied for that 5-10 minutes before the session started.

Carlos: Right, I guess there is that. But then once we get in there I felt a bit more comfortable. You know, you’ve given that session several times and that’s kind of when the habit start to kick in. I was happy. One thing I was happy about with the session is that we were able to still had some conversation. One of the things that I don’t particularly, I should say care for. My style I guess that’s a better word. I like my style to be a more interactive. I feel like if I’m just sitting there the only one talking in a presentation then I’m doing something wrong. I was glad that we were able to get some feedbacks, really some comments, right and kind of start a little bit of dialogue there.

Steve: Yeah, with that it almost seems like the larger the room is, the more difficult it is to get people to interact because they are afraid to speak out in the bigger room.

Carlos: Exactly right. That’s right and I guess, to that point we had. The room is, it wasn’t that humongous room because some of those rooms were big and I don’t remember exactly how many people would fit in there but we probably had, I don’t know, I’m going to say 50 people in there but I guess that was just on the cusp of still people being able to interact with each other. Once you get more than that it is like that Backup and Recovery session for example would have been a lot more difficult to do, I think.

Steve: It was a good session and I think it was the end of the conference session which is a tough timeslot but it went really well and I enjoyed it.

Carlos: Yeah, well thanks.

Steve: I thought one of the things I caught that’s really interesting is you’re talking about different products and tools and you bring up DBA Tools. I forget exactly what feature you mentioned but something with DBA Tools that you demoed and then Andy Levy, afterwards comes up or he said something about it because he works on it, and afterwards he came up and he said he actually worked on that thing, that feature that you were showing. That’s right so he had check in code. To him it was kind of a cool thing that the code that he has worked on has been shown at Summit. Only the best for companeros, right? So kind of cool to be able to show work and then that was it. We kind of went home. We went out to dinner with Derek and I headed to the airport.

Steve: Yes, so unfortunate story at dinner. We made a new rule that anyone who chokes and needs the Heimlich Maneuver has to buy dinner.

Carlos: Oh man, that was a bit scary.

Steve: We sat down for burgers and onion rings and I took the first bite of an onion ring. It wasn’t that big of a bite and somehow it got lunged in my throat. I wasn’t breathing. I did not have a working airway at that point.

Carlos: It was blocked. I was talking with Derek and you started kind of making a noise. I looked at you and your face was read and I was like, “Hey, are you ok?” I kind of pat you on the back and looked at your throat and I was like, “Oh man, we have a problem.” And we were in a booth which made it worse, we’re sitting next to each other and I was actually surprised that how quickly you were able to get out of that booth.

Steve: I wasn’t breathing. I had an extreme motivation at that point because I had an onion ring lunged in my throat.

Carlos: Yes, so that boy scout training coming back. That was a bit of a scary moment.

Steve: Yes, so very special thanks on that one, Carlos. I really appreciated that and I ended up buying dinner because of it.

Carlos: We’re glad that you’re still here and we hope that we never have to invoke that rule at dinner again.

Steve: Yes, I hope that it never happens again either, to me or anyone I’m dining with.

Carlos: Yes, that’s right.

Steve: But what a way to finish off a week.

Carlos: That’s right with a bang, that’s for sure. Ultimately that was our conference. We appreciate you companeros for sharing those of you who were there. For those of who weren’t, we’ll hope to connect with you at some other time.

Steve: Yes, and maybe next year at our SQL Trail Mix.

Carlos: That’s right, yeah. I think that’s going to do it for today’s episode. Our music for SQL Server in the News is by Mansardian used under Creative Comments. You can reach the shownotes for today’s episode at sqldatapartners.com/summit.

Steve: Or at sqldatapartners.com/117.

Carlos: And as always if you have thoughts about what you want us to be talking about on the podcast, you can reach out to us on various social media methods. One of them being LinkedIn. You can reach out to me I’m Carlos L Chacon.

Steve: You can get me on LinkedIn as Steve Stedman and we’ll see you on the SQL Trail.

Episode 116: Are people still using PSSDiag?

If you have ever worked with Microsoft support, you probably came across the PssDiag tool they use to collect information about your environment.  My experience with the tool is it always seemed, well–a bit clunky.  Now, it does collect lots of information and seemed to be a bit overwhelming when I was first a DBA.  Fast forward a few years and PssDiag was not a tool I use anymore and was not sure if anyone outside of Microsoft was using it.  I run into Jared Poche in Raleigh and he mentions he really likes the tool and mentions some of the improvements they have made over the years.  Well, you can guess where this is heading.

We talk with Jared about his experience with the tool and cover some of the basics about how you might go about using this tool in your environment.  There are two pieces to this tool, the first is the utility that collects the data and the second is the reporting piece SQL Nexus.

Are you using PssDiag?  Let us know if the comments below.

Episode Quotes

“It’s a really good tool in that it captures data from a lot of different sources and you tend not to miss much with PSSDIAG.”

“It is absolutely the go to tool for Microsoft.”

“Again using PSSDIAG and the Nexus will give you very visible data.”

“My biggest concern with PSSDIAG is people being too aggressive with the data they are trying to return from it.”

“Running it over the network is a really bad idea.”

Listen to Learn

02:47   What is PSSDIAG? What it runs and its usage?
04:54   Is Microsoft’s PSSDIAG and the publicly available different?
05:38   Using PSSDIAG in a small shop or in a large shop.
06:51   SQL Nexus
08:19   Improvements on PSSDIAG
09:42   How often Jared uses PSSDIAG?
10:13   Compelling reasons why to use PSSDIAG
16:04   Hand off point of getting information about the system, transitioning to smaller data set
20:52   PSSDIAG or DMVs? Which is a great place to start with baselining
24:12   Other common parameters in PSSDIAG, light capture vs. detailed capture
28:14   Tips and suggestions for starters
32:02   SQL Family questions

About Jared Poche

Jared Poché began working with SQL Server as an instructor for certification classes and has a passion for teaching and performance troubleshooting. Jared spent 10 years providing customer support at Microsoft, most recently as a Sr. Support Escalation Engineer. He is currently a database engineer working in Morrisville, NC and is leveraging his extensive knowledge to develop online coursework for SQL Server.

Transcription: Are people still using PSSDiag?

Carlos: This is Carlos Chacon.

Steve: I’m Steve Stedman.

Jared: I’m Jared Poche.

Carlos: Jared, welcome to the program.

Jared: Thanks for having me.

Carlos: Yes, it’s good to have you on and we’re kind of taking a break from our usual suspects at ChannelAdvisor. We met at the SQL Saturday in Raleigh and we’re talking. Ultimately our conversation is on PSSDIAG, but always good to talk with the ChannelAdvisor folks and your team is growing. You’ve added a few new community members that I know of and that’s always fun.

Jared: Yeah, Anders spoke at the SQL Saturday in Charlotte recently. I think we have 5 people there.

Carlos: Ok, very good.

Jared: I’m still kind of astonished by how active the DBEs and DBAs are in the community.

Carlos: That’s right, and so companeros, for those listening and I go back to a comment that Brian made, one of the DBA Managers over there. He said, “Some of the folks at ChannelAdvisor are very bright people and you don’t hear about them in the community.” And while Jared you’re getting out there a bit more of late, I kind of feel like you may fit that mold. Lots of the companeros may not know you unless they are kind of in the Raleigh or North Carolina area but hopefully that will change particularly with this interview.

Jared: Yeah, I’m trying to fit that mold. Yes.

Steve: Yeah, and I think the podcast is definitely changing that for some of the ChannelAdvisor DBAs that have been on.

Carlos: That’s right.

Steve: I think ChannelAdvisor has the record now for the company that has had the most guest.

Carlos: Unless we’re going to count Microsoft. Microsoft may technically still be ahead with all the PM they’ve had on.

Steve: Oh sure, yeah.

Carlos: Other than Microsoft… those two different layers are way ahead of everybody else. So ultimately one of the conversations we had, again kind of going back to that Raleigh conversation, and we were talking about performance and collecting information about the database. I can’t remember exactly what your comment was but you mentioned PSSDIAG. And I think my comment was, “Ugh, does anybody still use that?” And you said, “Yes, it’s great.” And I thought, “Oh, we’ve got to have you on the program to talk a little bit about it.” And so for those who haven’t use that, I guess let’s do a little groundwork here. Let’s just talk about PSSDIAG, what it is, and maybe even a little bit of history of it.

Jared: Ok, so PSSDIAG is this fairly own composing tool that Microsoft has developed and makes publicly available. It’s up on, I can’t think about the site is currently, but it’s publicly available for download. And this own composing tool that is used primarily for SQL Server performance issues. There are a few other odd circumstances you might want to run this but mainly you use it for, and Microsoft encourages to use it to gather data about performance issues they are having with SQL Server. And it runs a performance monitor output in the background. It runs a profile or trace on the latest version an xEvents trace if you want. It runs a bunch of queries to get information in DMVs. It will produce like 50 or 80 files of output in the folder that produces and it can produce quite a large volume of data. And the idea is you run this thing, you might run it for half an hour or whatever or if you had a different window when the problem occurs. And you know the problem might happen sometime in the next hour or so, you want to run it and capture data while whatever problem is going on. It’s a really good tool in that it captures data from a lot of different sources and you tend not to miss much with PSSDIAG. There are a few odd things occasionally that you might want to add on to say the normal configuration but it’s a really own composing tool and it puts a lot of data at your disposal.

Carlos: Right. Now, you are a former Microsoft employee, right?

Jared: I am. I was working at the CSS site in Charlotte for about 10 years. And a little bit of a contractor as well, a year and a half.

Carlos: I have heard, I guess maybe you can clarify for us. So traditionally you’re going to be exposed to PSSDIAG for the first time, at least I was. If you’re company has Microsoft support you call them up, you know, you’re looking at a performance issue. This is the go to tool, right, for them?

Jared: Yes, it is absolutely the go to tool for Microsoft.

Carlos: One of the things that I had heard was that if you go to their support site and what’s publicly available is slightly different, true or false?

Jared: They have an internal version of the tool, yes.

Carlos: Ok, but from the nuts and bolts perspective ultimately the same?

Jared: Ultimately the same. There might be some slight differences in some of the specifics of what options you have available. GitHub was what I was thinking of earlier. The current publicly available version is available on GitHub, and in the past there were some slight differences in kind of the custom configuration options you had in the internal tool and the publicly facing tool but they are 98% the same.

Carlos: Got you. Ok.

Steve: So then, I mean if you’re doing a support ticket with Microsoft they are going to ask you to run this. But if you are a DBA out there in a small shop and you’re having some performance issues, is this something that’s going to be easy to use and just that they could run and dig through the results and get some value out of it or is it more just for Microsoft to use?

Jared: It is something that you can use. As far as somebody who is in a smaller shop, I would actually have less concern about somebody in a smaller shop using it than somebody in a bigger shop with very big, very active, very busy servers. They don’t know what’s configured. The question that is being asked many times, “Will PSSDIAG affect my performance?” And the answer is possibly, and it depends on how you configure it and how busy your servers are. So in a smaller shop I’d be less worried about it. When I started using PSSDIAG in 2004 I think, I wasn’t very versed in performance troubleshooting at that time either and so I spend a lot of time looking at the perfmon manually. I’m looking for the profiler manually. There is also basically a sister tool with this which is SQL Nexus which is also available on GitHub. And what SQL Nexus does is basically grind through a lot of the data that PSSDIAG provides. Does some analytics on it, you know, figures out what’s the biggest resource consuming queries are. Your big wait types and puts it into a nice GUI set that you can see that stuff and consume it much more easily. You can select through it manually but SQL Nexus does make a good job of making that easier to understand.

Carlos: And worth noting in the beginning, I can remember being in that same state as far as just kind of getting my feet wet so to speak from all the tools that are available and taking a peek at that. One of the things that is a bit I’ll used the word overwhelming, or can be overwhelming of PSSDIAG is because it’s collecting all that data. It’s almost like where do you start.

Jared: It is difficult to know where to start.

Carlos: Yeah, and you kind of have to. I think if you have a good understanding of what all of those pieces are. It makes a little bit easier but starting there like your first troubleshooting starting place. Yeah, it’s going to take some time to get through, so be patient.

Jared: It can be intimidating at first. I do think that SQL Nexus makes a lot of that easier by kind of figuring out what’s important for you in doing a lot of the analysis by itself.

Carlos: Well now, let’s talk about some of the improvements. Again, so my experience with PSSDIAG is probably, gosh. You know, 7-8 years ago at least now. I remember SQL Nexus being a bit, I hate this word, clunky. But I felt like it was beyond maybe what I could do at that time. I felt like I didn’t have the experience to get it all set up. I never quite seem to fit that right. Is that getting better? Is that easier to use now?

Jared: For me it has gotten a lot better over the years. The last couple of releases have done more, there is more things exposed such as they added in something to get out the details of say sp_configure which is something that PSSDIAG has gathered for a long time but it will flag things in that that are unusual and things you might want to reconsider. I think that the GUI interface is better than it was. I think there is a lot to positive say for that tool, so I feel really good about SQL Nexus. And I kind of feel like I was doing myself a disservice because when it came out at first I was just kind of like, “Ahh, I don’t need that thing.” Because at that point I’ve been using PSSDIAG for a couple of years and I knew what to go and get and what to look for although it does in the long run it did saved me a lot of time using Nexus.

Carlos: I see, interesting.

Steve: How often do you end up using PSSDIAG in your job?
Jared: Currently I do not use it very often. I mean currently if I’m looking at a performance issue, and this is something that I started doing a lot when I was still working at Microsoft. A lot of times I’ll just use DMV queries and figure out what I can from that about an ongoing situation, and I feel like that gives me information very very quickly I can use to diagnose problems.

Steve: Ok, great.

Carlos: Now, again going back to our conversation. One of the things you got to mention was one of the things you thought PSSDIAG was really really good at. So we’ve talked about kind of collecting the metrics and whatnot, so what are some of the compelling reasons of why you use PSSDIAG? Other than just collecting a whole bunch of metrics, like you everything that you might need.

Jared: Well, it is all encompassing and doesn’t miss a lot so that’s a positive. The perfmon collector gets you everything you’re going to need in the vast majority of cases. The fact that it gets the information from the windows event viewer can really be helpful on occasion too especially if you’ve got a situation that you have disk issues. You might have errors that will come up and say the system log. That’s not something you’d necessarily think of unless you had a tool that’s helping to expose that information for you. One of the things that I got to really be fond of. In the latest version they have improved the ability get xEvent traces, so that’s something that’s been in the last couple of releases. But the last one that I saw before the current release, it was till kind of a stub, like there was a tab for configuring profiler and you really want to get away from profiler. I wouldn’t even think about running PSSDIAG with profiler on the servers on ChannelAdvisor in production. There’s no way.

Carlos: Interesting. I guess I want to come back to that, right.

Jared: Well, that’s a matter of overhead. I mean using xEvent traces and I’m more familiar with looking at profiler traces but I’m getting pretty familiar with xEvent traces. And the xEvent has a lot less overhead and the overhead of profiler can affect the performance of our system. That’s the thing in the past. And the busier your server is, the more likely that is to be a problem. So the fact that they’ve improve it because the xEvent thing used to just be a stub. And actually I remember one of the releases a couple of years ago there was a tab for configuring an xEvent trace and if you try to use it it would say, “Wait, this isn’t actually working yet.” So now it’s fully functional, now it’s fully complementary or equivalent to what you could get out of profiler so that’s good. One of the things I really really like out of it is in the output file there’s, it names a lot of things based on the name and the system and then the version of SQL Server, so the file is really long. There’s one file that is the perf_stats_startup.out, and what the script does it has like three different sets of DMV queries that runs. And it does one every minute that will do things like getting information from DMOS wait stats and it has another one that run maybe every 30 seconds that will get information about memory consumption. But the main loop, the thing that it runs like every 5-10 seconds runs queries against dm_exec_requests to tell you what all the active queries are especially the ones that are waiting or causing waits in the system. It does a second query, that for all the things that it got with the first query. The second query will return the statement that’s being run and the procedure it’s a part of which is very helpful for narrowing down what you want to look at and finding the section of code you actually care about. And the third query that tells you what the head blocker is if any blocking chains you might have. One of the things I would do at times to diagnos the situation very quickly is I would just open up that file, look at the information we’re getting from dm_exec_requests about active queries. I would have that open in Notepad++ or Textpad and I would just scan through that very quickly. Because it will run these queries like, I don’t know it it’s every 5 or 10 seconds. But each one you’re just getting one snapshot of what you’re running queries and what your waits look like. But if you see one snapshot, and another and another and you can thong through them all really quickly, it’s easy to come up with an overall impression of what’s going on. And to say, “Oh well, yeah I’ve got some I/O waits here and there but that’s not a big of a deal. But I’ve got cxpacket waits lined up for days. I need to figure out what’s going on with that.”

Carlos: That’s interesting as you mentioned that and I agree that that idea of the snapshot is very important. One of the things that I actually will have done at the PASS Summit is talking about baselines. So how do you, I guess thoughts around leveraging a tool like this which is kind of a snapshot but in a sense to give you, you want a broader than just what’s my current CPU, right? You want a little bit more history than that but it’s not something that you’re going to be running for weeks at a time either.

Jared: Yes. I would really consider using Nexus with that because of the fact that you can gather PSSDIAG whatever time you want to. If you want to gather an hour or two worth of data at whatever time of the day you feel will be much appropriate, go ahead and do that. But then use Nexus with that and you’ve got a very visible, you’ve taken that data set and made it very visible and very easy to look at and to refer back to it later because it’s like you’ve got all the perfmon data is in there, and average out. And you can say, “Ok, for these two hours this is my CPU consumption. This is what disks look like. This is how much free memory I had. Here’s what my main queries were, here’s what my mian wait types were, here’s what my resource consumption per procedure look like and you can have that kind of information for multiple sources and have it easily accessible through SQL Nexus.

Carlos: Ok, so you mentioned you kind of switch off of it, so when do you feel like it’s the hand off point? It seems like it’s a very handy tool particularly when you’re, I don’t want to say going in blind, but again if you think about the origin, right? So I am a support person, somebody calls me up, I don’t know who they are, I’ve never seen the system before, I want to get as much information as possible. Where do you see the hand off or the evolution perhaps of I need to get information about the system even if whether I’m new or whatever. I’ve just been tasked to this database to using PSSDIAG to, “Ok, now I know something about it. Maybe I don’t need to use all the pieces. I am more comfortable with the DMVs or something like that.”

Jared: Well, I think the more that you know the environment that you’re dealing with; the easier it is to transition to using a smaller data set. I think part of the reason PSSDIAG is as usual it is for say a Microsoft engineer is they don’t know what you environment looks like, and they don’t necessary know what they’re looking for. They don’t necessarily know what say unusual sp_configure settings you had. There is a lot of things that could potentially be the problem and there are more things that they might need you to consider. Now, in my case, part of the reason I made the transition, started doing more of my troubleshooting with DMVs at least were possible and to me there are still circumstances you just can’t replace PSSDIAG because of some of the things that it does. But part of the reason I made the transition was just for efficiency. So let me give you a hypothetical. Let’s say that you’re the customer and I’m working as Microsoft. We gather half an hour of data off of your very busy production server. You end up with say, three or four gigabytes worth of data in your output folder. You start zipping that up, you’re going to have to upload that to Microsoft. There’s a tool within Microsoft that when you upload that it will go ahead and unzip it and run it through the SQL Nexus output and I’ll eventually as the support engineer would get an email with basically the Nexus stuff already run. And the Nexus analysis that could take some time, a large data set, that can take a couple of hours. So I was actually on a critsit, if you’re familiar with the term. I was on a critsit with somebody in one case and they had a high CPU issue. He had been troubleshoot it himself and actually the problem went away before he contacted Microsoft and I was still working with Microsoft at that time. But the time I get on the phone with him the situation has been over for two hours. Now, PSSDIAG, I could configure PSSDIAG and send it to him and say, “Ok, if your problem happens again start running this and give me a call.” But I kind of wanted to see if I can do something more than that. I was thinking and I said, “So how did the problem go away? What happened?” And he told me, “Well, it stopped on its own. I was trying to figure out what was going on and it just stopped. I didn’t do anything.” So I asked the guy, “Ok, did you restart SQL or did you say flush the procedure cache?” Because that’s something a lot of people will do and it was kind of frustrating when you’re a Microsoft engineer. “Is the problem still going on?” “Oh no, I rebooted the server.” “Oh well, I guess I can’t gather any actual data.” But it hadn’t restarted, it hadn’t flush the procedure cache and so I thought to myself, “Well, whatever plans were consuming all the resources, consuming all the CPU, they should still be cached.” Let’s go run a query against dm_exec_query_stats that I had and I ran that and we looked at the output. And he had about 5-6 queries that were consuming significantly more CPU than anything else. Like you look at them in descending order and after about query 5 the CPU dropped by three digits. That’s actually unusual very often when I see a high CPU case and I’m looking on the DMVs. You’ll see that there is one or two things that’s really killing it. But in this case, I used the DMVs, I’m able to see the cache plans. We can look at the cache plans. We can go find the procedure that is related to it. We look at the table, we look at the indexes. I was able to get to everything I needed just from running a query basically against dm_exec_query_stats. And I didn’t have to have him gather data and upload it to me and I spent hours waiting for the data and then analyzing the data and then get back to him. I had an idea very quickly of what was going on. And after that one, that made a really compelling argument to me about the efficiency of using the DMVs.

Steve: Ok, so then just sort of comparing that efficiency and then getting all the information there. I mean, if you’ve got a new server that you are now responsible for you’re going to take over maintenance on. Would you say that, I mean, PSSDIAG would be a great place to start to get that sort of baseline and so that you know what’s going on with it and then later deal with the DMVs or you just sort of skipping it sometimes and go directly to the DMVs.

Jared: I think you could do either approach? Again using PSSDIAG and the Nexus will give you very visible data. I think the way it presents things might be easier for some people to see and understand and realize what normal and how this is different from that. But with the DMVs I think one thing that would be effective would be to have ongoing normal processes where you collect information from say dm_exec_query_stats or dm_exec_request and put that into a permanent table that you’re willing to keep for x amount of days. We some of this at ChannelAdvisor to keep track of say dm_exec_request information so that we can see waits that are going on inside the system and we can look back at a given period of time and see what kind of waits we normally see on a Wednesday at 2 o’clock in the afternoon. We can see that. You could do either approach. Like I said I like Nexus for the fact that it has a decent GUI on it. But you could also just take the DMVs have automated things to gather that sort of information. It depends on how long you want to retain that information. You have to be a little careful about trimming your tables if you’re taking that raw DMV information and storing it somewhere as opposed to say running the PSSDIAG and running the Nexus. Now you have the nexus output, you can actually get rid of the PSSDIAG data, however any gigabytes that is. You can also, Nexus takes the data and puts it into a database and it actually needs a database within SQL server that can connect to and juggle the numbers around. But you can backup that database and then at whatever point you can move it to another server. You can point SQL Nexus at it and it’s very effective in that fashion. Ok so for baselining to me you could use either one. It just depends on your preferences.

Carlos: So we talked about wanted to hit some of the new features and I think you’ve mentioned a couple of them, obviously Nexus being one. So as far as like, I’m sure even the documentation has improved a bit so you mentioned some of the parameters or settings you could have. So if I just run PSSDIAG with no parameters I’m going to get the event log, I’m going to get the SQL Server log, I’m going to get the profiler trace. I am assuming it’s still the default. You mentioned I could get xEvents, now I’m assuming parameter change and then I’m going to get the performance monitor metrics. So what other common parameters am I using when I am talking about PSSDIAG?

Jared: Well, the thing currently is and this is where people go wrong with it or can go wrong with it is now there is basically default packages or default scenarios or scenarios that you can just basically say I want to gather general performance or I want to gather detailed performance. And one of the things that I would suggest people not to do with PSSDIAG especially on the production environment would be to go in and start by telling it to do a detailed performance output before you’re familiar with the tool and have an idea of how much data it’s going to capture in your environment. I would go with a general or light option. The big difference between a general or a light capture and doing a detailed capture is that when you do a light capture you will gather the events for whether you’re on profiler or xEvents, you will get RPC started and completed and you will get SQL Batch started and completed. And you will get errors and you will get a bunch of other things as well but the main things will be interested is batches and stored procedures started and completed. When you go to detailed, you’re going to gather a lot more events because you’re going to gather statement level events. You’re going to gather every statement starting and completed within SQL Server and you’re going to gather execution plans of whatever type you’ve selected or whatever type is the default for the detailed package currently. And think about it, if you got a store procedure and that store procedure has 50 statements in it then by going to a detailed trace you’re going to gather, instead of gathering two events for that stored procedure – starting and stopping, you’re going to gather those two plus 50 for each statement starting, 50 more for each statement stopping and for each non-trivial statement that you’re going to capture the execution plan. So there is a big difference in terms of how much data that you’re going to be gathering and thus it’s going to be more likely to affect the performance of the system in some way, and it can. And if you’re SQL Server is pushing 2,000 transactions per second that’s probably not a problem. If it’s pushing 25,000 transactions per second the same package might bring that server to its knees. And especially if I’m not familiar with running PSSDIAG in my environment I would start off by doing a general or a doing a light and seeing how that works before I try to crank it up.

Steve: So I guess from that perspective it also makes sense if someone is just learning it if they have access to a test server that might have far less load to just give it a test there first to make sure you understand what do light versus the more detailed is going to do.

Jared: Absolutely. And there are some interesting circumstances like there are things in there that are made for gathering information about always on or gathering information specifically about memory issues or things like that. I think there is used to be one for, like there is still a collector for service broker and there are circumstances you might want to gather those things because it provides a lot of information. There is one for link servers that will run all the queries to get information about what link servers you have set up. So there are a number of useful little things like that. But my biggest concern with PSSDIAG is people being too aggressive with the data they are trying to return from it because it can affect performance. I had cases when I was working at Microsoft that I would go in with the PSSDIAG package and we’d run it and we gather data and I analyze the data and realized, “You know what I can’t use this because all I can see is the SQL trace waits.” I can’t tell what else is causing problems on the system because the tool is gathering too much information that were waiting for that information to be written out. And I can’t see any other bottlenecks because of this bottleneck.

Carlos: Ok, so I guess then at this point for someone who is about to go and give it a shot maybe after listening to the podcast. I mean, besides keeping the sampling light there. Any other tips or suggestions to help someone get started?

Jared: One or two that’s big. Another thing that affects performance is running this over the network as oppose to running it locally on the server. Running it over the network is a really bad idea. I would dissuade you from doing this if it all possible. The idea is that SQL has these delays when it’s trying to write the information out. Now, PSSDIAG does server side trace but if you’re going to be using profiler or using some other tool to do your tracing, I would really suggest not doing it remotely. You can run PSSDIAG remotely and tell it to connect to this server and this name but it introduces a much longer latency in terms of how long it takes for the trace to write out that data because now you’re… And that is affecting how the SQL Server engine operates now, so that’s a whole different magnitude. The other thing that I’ve seen that causes problems and it’s not specifically with PSSDIAG is not to run a whole lot of traces at one time whether you’re using xEvents or profiler. There used to be commercial tools. I won’t name any, and I probably couldn’t remember them all of them. But that would run four or five or more profiler traces at a time and that would affect the performance of the system because SQL Server is trying to write out things and keep track of things for each one of those different trace files. Especially when you’re doing multiple and you’re doing it over the network, that can really be consequential.

Steve: Yup, interesting. Well, ok so a side note put on that running profiler traces. I mean, what would your take be if someone wanted to run a profiler trace like 24/7/365, all day, all year just to keep track of every query that’s ever being run? How big of an impact something like that can have on performance?

Jared: Again it would depend on how busy the server is. On a busy production environment I would ask them to find exactly what it is they are trying to find out here because I don’t think that’s what they are trying to find out. I think that’s a lot more information than you could reasonably use. How would you aggregate that information? What are planning to do with it? And in any case I would stay away from using profiler at this point in any case because xEvents is now a super set. It has been since SQL 2012, has been a super set of what profiler can do. We have more kinds of events of xEvents now. It provides more information to you and it’s more efficient. It was designed from the ground up to be faster and to have less overhead than profiler. I really encourage you. If you’re more familiar with profiler, I entirely understand. But I would really go ahead and pay that price because I think it’s more than worth it.

Carlos: Well now I’m interested to find out what happens with profiler. I don’t think it’s going away and the reason I say that is that they just announced. Or maybe not just but one of the things we talked about SQL Server in the News is they’ve added query store as a source for profiler.

Steve: Alright, shall we move in to the SQL Family section?

Jared: Sure, go for it.

Carlos: So Jared, how did you first get started with SQL Server?

Jared: I got started when I was a part-time community college instructor. I had been teaching for one semester and my older sister actually work at one of the colleges I was working at and she asked me if I knew anything about databases and my answer was no. At that point the college and this was Forsyth Tech in Winston, Salem. They did some certification classes for SQL Server and one of them has actually in one of their 2-year degree. And they needed somebody to teach this classes for the next semester, so she came to me and asked if I could do that and I said, “Ok, I’ll look into it.” And I basically picked up a book and taught myself enough of SQL Server to get certified in it and then I taught the classes there for about three years. And from there at that point, the interesting thing about teaching is whenever you’re trying to teach people something you have to know it to a greater degree than your teaching. And also they’ll ask you questions about the subject that you’ll never think to ask yourself. So it’s like you as a teacher end up learning even more from having students because you’ve already figured out all the questions you had in your head but now you’ll get a whole bunch more questions and you have to figure out how to answer them too. So I taught that for about three years and that was when I applied for this helpdesk e-job that I saw on Monster which actually turned out to be a contract position to go work at Microsoft at Charlotte.

Carlos: Interesting. So it’s very similar almost to like getting started with presenting in a sense, right?

Jared: Yeah.

Steve: So if you could change one thing about SQL Server, what would it be and why?

Jared: That’s actually pretty difficult. Trying to think of, you know, I’ve had pain points here and there over the time. I’ve seen people do some horrible things with service broker. I’m trying to think, what else? At the moment I guess the biggest thing I would come up with is page latch waits because there is such a big consideration and a kind of a thorn in the side currently at work so I guess I would try to some find way to eliminate that level of pessimistic concurrency. We’re trying to move to doing a lot of things doing in-memory optimized tables and whatnot to move to more optimistic concurrency because with our throughput, with our load, we need to have that. We need to get rid of the contention we get from page latch waits. That’s a big consideration.

Carlos: Got you.

Steve: Ok, very nice.

Carlos: What’s the best piece of career advice you’ve ever received?

Jared: Difficult to say, I think one of the things I would go back to is when I was still contracting, when I was still starting off at Microsoft. I was speaking with one, my tech lead at that time was Bill Carroll. He is now an Escalation Engineer. He’s been at Microsoft forever. I was working on some performance case and at that point hadn’t a lot of experience working. I had no experience before I came to Microsoft working on performance cases. I was finding them difficult and Bill Carroll said, “We will always have room for somebody who knows how to handle performance cases around here.” And so that I think really probably was the best piece of career advice because at that point it gave me a niche. It gave me something to figure out and try to focus on and I’ve done a lot with that so I’ll give Bill Carroll the nod there.

Carlos: There you go.

Steve: Alright, and our final question. If you could have one superhero power what would it be and why would you want it?

Jared: That’s funny to me in part because I’m an avid video gamer and my wife and I used to play City of Heroes so we have played superheroes many times. For selfish reasons I would like to say either flight or teleportation because that would really be cool. But honestly in real life if I could have a super power that would be the power to heal other people just because who I am as a person if I could do something that could make life better for somebody else. If I could take away something horrible that’s happened to them that would be a great thing to do.

Carlos: Very cool. Well, awesome.

Steve: Alright.

Carlos: Well Jared, thanks so much for being with us today. We appreciate the conversation.

Jared: Well, thanks for having me.

Steve: Yeah, that was great to be able to chat with you, Jared.

Jared: Same here.

Episode 115: The Companero Conference Retrospective

Companero ConferenceIt was a beautiful idea.  Gather together with a group of our podcast listeners in a conference type format and find ways for us to connect and help each grow.  I had seen it down in other events; however, I wanted to add my own flavor.  I wanted this to be different.  While the event turned out great and we got good marks from the attendees, we didn’t have the response we were hoping for.  In this episode of the podcast, we have our companero conference retrospective.

We try to give our honest feedback about what we did well, what could have been improved, and what the attendees reported.  I can’t say enough about the help I got from our speakers–Jonathan Stewart, Kevin Feasel, Randolph West, Tracy Boggiano, and Doug Purnell.  They were awesome and the event was much simpler because of them and their commitment.  They are truly my companeros on the SQL trail.

WHAT SHOULD WE DO WITH OUR CONFERENCE?

We would like to send you a survey on what you think the future of the conference should be and whether we should keep it around.  We look forward to hearing from you.

Episode Quotes

“I wanted a way that people could come together, could talk with each other, connect, build those relationships.”

“We felt like we were able to create that intimate environment where people could participate.”

“You’re not just a face in the crowd. You’re actually there participating and I think that’s the difference.”

“I think the pull of those bigger conferences was just too big and lot of people ended up going to either one of them.”

“I don’t really like the word conference… because we’re going for the unconference idea.”

“I think we learned a lot. I think I’m glad that we went through it and thanks obviously to those who attended and to the speakers.”

Listen to Learn

00:39   Companero Shoutouts
07:25   SQL Server in the News: Import Flat File Wizard
09:06   SQL Server in the News: Service Packs to Cumulative Updates (Pros and Cons)
13:45   Links of the show notes
13:57   The reason/objective of putting up the Companero Conference
17:59   Challenges that happened
20:57   Conference outcomes – Are the conference objectives met?
25:04   What are the big challenges?
36:23   Thoughts or changes for future events, to do it again or not in 2018?
45:30   What to look forward to? What’s next?

Import Flat File Wizard

Transcription: Companero Conference Retrospective

Carlos: Companeros, welcome to episode 115. It’s good to have you on the SQL trail today.

Steve: Yes it is and it’s good to see you Carlos.

Carlos: Yes, and it will be actually good to see you in a couple of weeks here we go on to Summit. We don’t get to see each other as often as we would like.

Steve: Is this where we let out that we’re not actually in a recording studio.

Carlos: Yeah, oh boy! Was that a secret that we’re keeping there?

Steve: Yes, yes.

Carlos: Yes, in the MGM Studios or on Park Avenue or whatever it is, or the MBC Studios.

Steve: Yeah, so we’ll actually be seeing each other in a couple weeks as well as many of our companeros who maybe attending PASS Summit as well.

Carlos: Yes, that’s right, and we do have an event that will make a little bit more sense which we will talk about at the end of this episode. So today, we actually want to do a Companero Conference retrospective. We’ve been kind of evolving this idea of the conference right in front of our podcast audience and we went ahead and had the conference this October, and so we want to talk a little bit about it. How it went? Maybe some of the lessons we learned from it, the challenges that we had and then what if in the future it might have.

Steve: Yes, but before we do that. Do we have any companero shoutouts this week?

Carlos: We do. I want to give a shoutout to Kira, and you’ll forgive me but I don’t have her last name but the organizer of the girlsanddata.org site. She had put this together. I thought it was interesting in a way because there’s like girls who code at code.org, lots of information about coding and teaching kids how to code. And she felt there were lots of opportunities in technology particularly in data that you don’t need to be a coder, you don’t need coding skills to do. So she connected with Mindy Curnutt of Dallas. And Mindy has taken the ball running with this and they were kind of just in the Milwaukee area, kind of doing things locally, and Mindy has connected with them and they’ve done events now in Dallas and in Charlotte this last weekend. And I went down there for the SQL Saturday actually for the purpose of putting two of my girls attend the event. They had a great time, I thought it was phenomenal what they did. You know, my girls are not, they are kind of that age where they are resisting anything that dad tries to get them to do. So there’s a bit of that, right, going on they have that to overcome. And so my oldest daughter, I should say Andy Leonard’s daughter, if you want to hear really great things you should go hear Andy Leonard’s daughter. She was absolutely over the moon with it. My daughter is responsible as well it didn’t blow me to death. And so which may not sound very good but some of the languages she’s been using lately, that was a pretty much as good as it gets.

Steve: Ok, and what age is she at?

Carlos: She’s actually 15. The girls at data.org is actually directed more middle school girls, so the 6th to 8th Grade girls. She is a bit older than that but I thought, “You know what, I’m just going to put her in there,” regardless of what I think it will be. It wouldn’t be that bad. May other daughter is 13. She is an 8th grader currently, and again, I think she enjoyed it. But some of that feedback was hard to get because she couldn’t say. She wasn’t super negative about it but I think she couldn’t be super positive about it because for fear that I would sign her up for something else.

Steve: Interesting. You know what I think is really cool about that program is that, I mean there are so many different programs out there to try and introduce younger people to programming. Like you said, data can be very different than programming. I think that in this school system as well as like community college and university even further up on that level the education on data is usually pretty weak compared to the education in programming.

Carlos: Right, exactly. And so I think it was an interesting that she put together, you know, focused on that. They are using tableau as their reporting source and they have several pieces of information they put together and do different things there. And so I thought it was very interesting. They had an interesting mix of people, so in that class they were 15 or 17 girls, probably nearly half of which parents were attending SQL Saturday. They had actually made the announcement to several of the middle schools and middle girls camp so other folks weren’t currently tied to SQL Server were still attending that. So I thought that was good.

Steve: Alright. Can we just pause for a second I need to turn off some background noise.

[PAUSE]

Alright, I’m back. Sorry about that. Hey Carlos, I’m back, sorry about that.

Carlos: Not a problem.

Steve: And out next companero shoutout is from Eduardo Cervantes.

Carlos: Yes, Eduardo on the West Coast reaching out to us and reminding us that some of our listeners are developers. He mentioned, “I do most of development. I appreciated the podcast”, and connecting him with new features and things. So thanks Eduardo for reaching out and for also just responding to something on messenger. I’m sure none of our companeros are doing it but I have people reach out to me from time to time they want to connect on LinkedIn. I’m making a personal habit to respond. It’s pretty generic but you know it’s, “Hey, thanks for connecting. I hope you’re well. Let’s chat, like what do you need help with?” It kind of bothers me to no end when I just don’t receive a reply. Even if it’s just a, “Hey thanks” or something, right? Let me know that you’re there, that you have a pulse. So with that I guess a little SQL Server in the News is on file.

Steve: Yeah, so what’s up with this import flat file wizard you mentioned?

Carlos: Yes, so one of the new features they’ve put in and let’s really just. I don’t want to say dumbing down but they have the data import option so kind of using it in the background with SSIS like package that will get import data. So what they’ve done is they kind of slim that down a little bit and made that specific to flat file data so CSV and text file data. What it helps people do is to walkthrough the wizard because before on the import data you had to go in and kind of like, I can’t remember exactly what the button was, but to see the conversion with the columns are going to be. You don’t have to drill down a bit anymore. It will do those conversions. So first you got a preview of the data actually, so again very similar to features that are in SSIS. It’s going to import that data. You’re going to get to look at it, make sure the columns are all lined up. The next screen is, hey I had to change the column types based on what the wizard is telling you it thinks it should be and then you can import. Yeah, I think it’s just a little bit easier and make it a little bit faster. Do those little one offs since like we’re always importing just random Excel data for whatever reason, look up data and things like that.

Steve: Oh yeah, and there are so many different ways of doing it I think. And that’s just one of them that makes it easier so that’s great. Another SQL Server in the News items that we have is around service packs.

Carlos: It got very loud on your side or something.

Steve: Hold on a second, Carlos.

Carlos: Sure.

Steve: Alright, I’m back. Sorry about that.

So Julien let’s cut that out we’ll just go back when we start talking about service packs. I’ll explain afterwards what’s happening, Carlos.

Alright, so another SQL Server in the News item we have is on service packs. This came out a couple of weeks ago with SQL Server 2017. Around that time they also announced they are going to change the way that service packs work. Instead of doing service packs they are just going to go straight to cumulative updates. And for the first year, it sounds like the plan is on SQL Server going forward they’ll have… They are not going to have a service pack, they are actually going to have a cumulative update every month. So you get 12 cumulative updates and about at the end of the year those CU12 should be like the equivalent of what would normally be SP1 in the past. I think that there are some interesting takes around that because one it sounds like we’re going to, and this actually happened for that last couple of years now, is that cumulative updates have been as robust as service packs. Didn’t used to be that way years ago. But I think it gets away from service packs having sort of people to wait, I’ll wait for SP1 or I’ll wait for SP2 before I upgraded that version. Now there will be no SP so maybe people will say, “I’ll wait till CU1 or CU2” but who knows.

Carlos: Yeah, we’ll see interesting I think. I feel like this is kind of the dev ops coming, I mean not to the database perspective but that dev ops mentality making its way into applications. Obviously it’s kind of everywhere people are talking about it and now it’s just making itself involve or connecting with what we do on a regular basis. And so more applications like this will kind of go this route.

Steve: And I think it really highlights the importance of the software insurance.

Carlos: Oh yeah. We talked about that before we started recording and I think with this rapid updates, again, people are already talking about SQL Server 2018. It is interesting to see what will happen. I feel like, again, kind of with the Azure mentality of everything being in the cloud then releasing software so quickly that you’re going to almost have to have software insurance just because have you know when to upgrade anymore, you know just pick a year. List some feature that you’re really really interested in kind of those rolling updates. Yeah, it will be interesting to see what happens, what enterprises do with. So let’s just say in 5 years, and I’m making some assumptions here, but let’s just say there is a release of SQL Server every year through 2020. What do you there? It used to be like upgrade so like a big deal like you plan time every couple of years. Now I think we’re going to see a lot more versions out there than we ever have before.

Steve: Right, yes, which I think has some pros and has some cons. I think for the people who stay up to date they’re going to have more work to keep up to date there. And then I think for the people who don’t keep up to date, well you’re going to get outdated I think much quicker.

Carlos: Much quickly, yeah that’s right. So the gap is going to get a bit harder. But to their defense, Microsoft has been very good about the backwards compatibility and I think that’s even getting better.

Steve: Oh yeah, and I think hopefully with the more frequent versions it will make it easier for people to jump into the next version because you don’t have 5 years of changes or even 2 years of changes that you have to worry about compatibility with you’re dealing with once a year.

Carlos: Exactly. Yup.

Steve: Alright.

Carlos: So some interesting stuff. Ok, so our show notes for today’s episode will be at sqldatapartners.com/conference.

Steve: Or at sqldatapartners.com/115 for the episode number.

Carlos: So ultimately what we wanted to do today is talk a little bit about, and I guess have that retrospective for the conference that we put together. I feel like we should start with the premise of why we wanted to put this together. Admittedly, I was heavily influenced by my own experience, so getting involved with SQL Saturday, with the User Group, going to SQL Cruise, going to Summit. So 2013 was the first year I had gone to the Summit. I just had different experience then with other conferences outside of the community events. I felt like there was a way to continue that idea of having people participate more in the conference. This kind of goes back to Allen White, so this is all Allen White’s fault. With that idea that we all have something that we can share with each other and those traditional conferences, in my mind they do a great job of expelling information but as from a collaboration perspective and from how this apply to my own unique perspective, they don’t do a great job. And so I wanted a way that people could come together, could talk with each other, connect, build those relationships they could take home with them, things like that.

Steve: Yup, and that’s the key, the connecting and the relationships. I mean we stressed that as one of the most important factors in the conference from day one.

Carlos: Right. What it turns out is a little bit of a tough sell to a wider audience. I think as technology people they, I mean like they behind the screen and just consuming information. And so I think it takes a little bit of a challenge. It’s a bit of a challenge to reach out and say, “Hey, my name is Carlos and want to connect.” And that’s something that we’re kind of trying to work through. And what are the other things that I thought? Again, I’ll reference SQL Cruise which is now Tech Outbound. I think one of the challenges they had was just being on a cruise itself. A lot of people that I talked into, I would talk about it and then they’ll like be, “Well, it’s on a cruise.” And then they do like, “Nope.”

Steve: Oh yeah. I remember hearing that on many occasions like where I wanted to go and pitch it with the company I worked for at that point in time and they are like, “What? You want us to pay for a cruise?”

Carlos: Right, exactly, right? So they think there’s a bit of a challenge there but some of the concepts were really good. One of the things that I miss or I lack about the SQL Cruise experience was that ability to kind of give back, to give my own ideas or thoughts around some of the topics. And so the rise of the unconference has played into this idea and I actually really like the premise of this idea of, ok we’ll come up with a track so here are the things that we can talk about. But then you know what we’re actually going to decide what we do talk about when we get there and then play it accordingly. And so that idea resonated really strongly with me and so that was what we want to put together.

Steve: Yeah, and I think that, I mean we went in that direction and I think that really is what we ended up putting together.

Carlos: Yeah, so I guess maybe let’s just jump to those outcomes. I feel like we were able to meet our objectives. We’ll talk about the challenges. We ended up having as we had 8 attendees, we had 5 speakers, myself and my wife, so there’s 15 of us in total. And there was some concern about the size and we went, we kind of had this thug of war as to whether we were or weren’t going to have it. And then really it came down to the speakers. I reached to Kevin, Jonathan, Randolph and Tracy, and I said, “Look guys, this is where we are. Do you still want to do this thing?” And they said, yes, and I’m like, “Ok, let’s do it.”

Steve: Well, and that’s one of those were I think to speaking of reaching out speakers. I mean, I was originally on the list as one of the speakers there. And in doing that I’d plan on attending but then when we got to the point where we’re sort of making the go, no go on whether we’re going to proceed with the conference or not it came down to just the overall amount of workload we had combined with the cost in getting there – coast to coast flight. That’s when we realized that we don’t have enough people showing up to be able to break even on it so that’s when I decided to opt out at that point and hopefully next time we can be in a position that it makes sense for me to be there.

Carlos: Yes, so that’s challenges from us as partners, only having two people hold down the fort. We have a project that needed some attention so that’s also a role. And then that week we ended up having two corruption issues with companies that we had not previously done work with and so that made for a very very challenging schedule.

Steve: Yup. So Wednesday, that day of the conference, that first day of the conference, that was the day that I spent pretty much the entire day fixing corruption for new client. And it was one of those things that had we both in there and hadn’t anyone left to cover anything. We would just have to say sorry we can’t do that we’ll have to help you another time would not been a good answer for that client.

Carlos: Yeah, that’s right. So that’s still something that we have to figure out from our end. But I play it to the reason for only having one of us there. But I’ll just go back to the objectives for a second. We felt like we were able to create that intimate environment where people could participate, get to know one another, and I think there were two major reasons for that. One, again because of our numbers we ended sitting in kind of a horseshoe shape, the traditional U. Attendees really like this because when someone started speaking even if that was a speaker which at the beginning it was mostly just speakers talking, so all the speakers were in there. They for the most part knew each other or at least of each other so there were some relationships already there. When people started speaking you don’t have to turn around to engage them. A slight tilt to the head you’re looking at that person and you could engage from there, so people really really like that idea. And then the other was the way we actually start the conference is that we started the conference with the four rules and then we did the three questions. So the four rules, some ground rules as to what you can expect out of the conference. And if you don’t have those things just speak up and then the next portion is just introducing yourself and actually having some questions to go off of. They are open ended to invite some discussion and to start making some connections. You can find who your people are. I think the combination of those two things helped to increase that engagement. Again, with the speakers being there, and making that commitment to be there for the two days we felt like we could make our objectives so that’s ultimately we went through with it. The other thing there and I think from an objective perspective is that the attendees were able to get to know each other, and we had great content. Right, from all of those things we felt like the objectives got met. I guess we didn’t feel like we were promising more than that and that’s why we were able to deliver. I was little bit nervous and I told the attendees beforehand that we’re going to have a small group. And even at 60 people that’s not a ton of people. I mean, when you think about other conferences that’s very very small comparison. But we’re going to be a smaller group and at first they were kind of like, “Huh, this is a little bit strange. I thought we had a few more.” But by the end I actually had two people say, “Hey this is right sized”, and we’ve actually again from a future perspective we’re going to actually come back off of that 60 number and it looks like maybe 30 is our new target. And again, we want everyone to be able to sit in that horseshoe shape for people to face each other basically with a desk and that kind of factors into that idea.

Steve: So you’re not just a face in the crowd you’re actually there participating and I think that’s the difference. And so then that’s was the other piece participating, so one of the ideas. We had a mix of traditional conference and unconference. So we had the speakers they each had a dedicated session that’s what we had promoted on the website as well. These are the things that they are going to talk about but we spice up that a bit and we actually solicited input and we said, “Hey, what it is that you guys want to talk about?” Everybody wrote those down on index cards. We put them up on the wall ad we actually had everyone vote on what they want to talk about. And one of the things that made it up unto the list was SSIS.

Steve: Interesting because that wasn’t even one of the things that we had anticipated that people would want to hear about.

Carlos: Exactly and so this was interesting for two reasons. It stemmed the discussion into two ways. One, so the card you set SSIS. We started with some questions and then we asked who wanted to talk about it, and one of the attendees, John. I think we mentioned him on last week’s podcast. John actually said hey I’d like to talk about that. So he and Doug actually lead that discussion and this is on Day 2 in the afternoon and because we already have that time together everyone had gone familiar with each other and so we were able to have, I don’t want to call it impromptu. It wasn’t as polish if you will again as a traditional session but it was very specific to what people wanted to talk about. Doug happen to have a VM, he pulled up his packages and we were able to do some demonstration and actually talk through, here’s what I do, here’s what we do and there were conversations all over the place and I learned some things just about logging that I didn’t know about. And then that turned into a discussion on [term unclear – 26:15] During all of that to which Randolph was like, oh I impressed. He hadn’t really seen it and he mentioned I don’t know that I would have spent as much time. Maybe it was only like a 30-minute discussion, right. But I don’t know I will spent 30 minutes on [term unclear – 26:32] But now that I’ve been exposed to it and we had in that context I fell like that’s something that I want to learn a bit more about. And so I really like that component of it. It was a very organic conversation that we didn’t have to worry about time necessarily and we kind of go with what people want to talk about.

Steve: Yup. Ok, so a lot of good outcomes there. As far as challenges what were the biggest challenges?

Carlos: Yes, so challenges, I think in the beginning we didn’t do a great job of getting user feedback. I think I know originally we kind of started with the idea of the lone DBA which I’m not opposed to. I think we knew we were looking for a segment of people we wanted people to easily identify whether or not they would come to the conference. And we have the lone DBA might be a good way to do that. Getting a hold of those people was a bit of a challenge and then even you companeros, right? We asked for feedback time and time again and we got not as much as we would have liked. You could also say, well, the way that we went about it wasn’t good. There were things to learn there. It felt like there was enough enough interest but we didn’t do a great job of connecting the dots and saying, “Hey, what it is that you guys really want to do and would be willing to pay for?”

Steve: And I think part of that, I mean, early on we were trying to get feedback there and I think that we just didn’t have the right venue for people to be able to present that feedback to us I think.

Carlos: You know, exactly that’s right. A lot of that initial feedback was I was contacting people directly and asking them, and I was reaching out to managers. I was very concerned about this idea of getting approval for people to come. And so I wanted to make sure that I could tear down from, and so when the manager looks at the website all their boxes would be checked. If they employee would like to come then they would get approval. And so from that perspective I think we did ok. I think maybe with that last mile of who should attend and why? I also think so having it in October so we knew that we were not Summit, and we are not Ignite, and we’re not you know. That’s not what we’re putting on even if our capacity 60 people we were not putting on that type of conference.

Steve: Right, that’s a very different type of conference what we’re going for.

Carlos: Oh yeah, exactly. However, having said that I don’t think that we fully realize the pull at least from the people in our network, right? So the people who are listeners, those are the two kind of we’re reaching out to, so speakers, SQL Saturday people as well. I don’t think that we had good data or the data that we had was wrong basically. We heard a lot of people saying, “Oh, I’m not going to Summit because it’s over Halloween.” We thought, ok, well maybe this is the year to try it. Even it’s not the same, we’re not saying it’s the same but let’s try to do something different. And I think the pull of those bigger conferences was just too big and lot of people ended up going to either one of them, with Ignite being the week before and then Summit being a couple of weeks later. The plans that they had in April and May changed and people ended up going or other things happened.

Steve: Right, and I think that’s a great sort of learning thing from being our first year doing it. How can we anticipate that back in April or May when we’re doing a lot of the planning? But I think now we’ve learned.

Carlos: And I think that, yes we talk a little bit about marketing, and so the other thing there is just a standardization. I think people have gotten very very comfortable with the way PASS us the events. Yeah, if you’re in that world already I think change is not, people seem resistant to change a little bit there. But I think one of the nice things is that we’ve had the event we can now have people, in respond to it give feedback on it. You can see it. Jonathan wrote up a very very nice post about it. You know, the speakers have come, they’ve seen it as well and so hopefully some of that will filter into the community.

Steve: The other challenge that we need to talk any about the financial side of it.

Carlos: Yeah, so that’s right. At the end of the day, a slightly different, so I guess this kind of goes into this standardization process. I’ll take a SQL Saturday haven’t been involved with those and putting those on particularly that it’s a free event. Because of the standardization that PASS has done, so PASS is sponsoring, Microsoft is now sponsoring, and you have all these vendors. It’s kind of a known quantity because they’ve put all that time and energy into explaining what it is. Having people go and whatnot. That from a sponsorship perspective, the entire event can run off of sponsorships and then lot of volunteers and all these kinds of things. So because we’re going with the SQL track model, we knew that sponsorship would be very very difficult because there wasn’t a traditional, hey you get to have a booth, talk to the attendees, raffle tickets, things like that. Let alone from the numbers perspective. And so that was a challenging idea as well. And so as a result we depended entirely on revenue from the attendee tickets. It ended up being about $4,000, our expenses were about $6,000, and that was another thing that’s bad decision. I knew that when we made that decision for Steve not to come out and to move ahead with it that we couldn’t cheat the people who had come and basically budget it, right? Because I didn’t want them to feel like they were getting short ended. I wanted to make sure that we were meeting the commitments we’ve made to them. But from a business perspective so as an entrepreneur you can’t contain to do that all that time and energy and then come out $2,000 in the whole doesn’t equate to long term success.
Steve: Now, if we were like a software development company and we have a whole bunch of products that were being sold out there and we could ride off that loss as an advertising expense perhaps or something, then yeah we can keep doing that indefinitely. But I think we really don’t have that. I mean, other than our consulting business we don’t have that other revenue stream.

Carlos: Or outside investment, things like that.

Steve: Yup, yup, so I think being able to set it up in way that it works so we don’t lose money to do it I think is a key factor to make this sustainable going into the future.

Carlos: Yeah, and so I guess we can talk about that future and I feel like we were pretty close. I mean, it’s not like we were at the Ritz-Carlton and doing all kinds of really crazy over the top things. Although we went on the cruise, we had a nice buffet dinner that was very very nice. We had plenty of food, the room was nice and all these things. It wasn’t over the top I think if we got from 8-16 we would have been in a better place. So I guess talking about the future, so a couple of takeaways. I guess another challenge. I knew it would be slight challenge to use the word companero. I don’t think I understood how challenging that was going to be.

Steve: And that’s one of those things that in retrospective it’s easy to see that now. I think just looking back at our podcast Episode 100 with Kevin slaughtering the name companero. Comp… however you say it in the beginning. I mean, maybe that should have been a clue but we kind of have fun with it. But yeah, I think that’s good point. It’s a tough one to say or even to spell for a lot of people. Even me.

Carlos: I don’t really like the word conference, right, particularly because we’re going for the unconference idea. I’d like to stay away from that word conference. And so some of the feedback that we got, some of the thoughts that we have about potentially putting on if we were to do an event in 2018, some of the things that we thought about. So one, renaming it to SQL Trail. SQL Trail, much easier, it’s kind of already part of our brand as well, right. “We’ll see you on the SQL trail”, we’ve said that on every episode since Episode 0. That has been part of the podcast. And so I feel like incorporating that would be a bit easier.

Steve: Oh yeah. And I’d be curious to hear what people think about that. I mean, does SQL Trail make more sense than Companero Conference? We’re always looking for feedback and those kind of things.

Carlos: That’s right. The reason I like it even and again kind of using, even SQL Cruise or SQL Saturday, they are not using that word conference. They have that short name that can be used to identify what it is without using the word conference. And so that’s what I would like to do, and so SQL Trail would be a potential.

Steve: Yup, I like it.

Carlos: The other change that we were looking for too going forward is we had the event for two days. What we did is we use, and it only ended up being about an hour but we did an hour of introduction. So maybe an hour and a half in the introduction, and then the session selections for the Thursday. Part of the feedback was, hey, let’s push that to either before the conference basically or separate it from the first day of the “conference”. And so what we were thinking of doing is making that 2½ days. So we will start for example on a Wednesday, at 2PM. You have the welcome, everybody comes, introduces themselves. You know, introducing that idea of the social into the start of the conference and then choosing sessions that afternoon and then having a dinner. Right, so then at the end of the day on Wednesday, you know the sessions, the schedule for Thursday and Friday. And then on Thursday and Friday you come and it’s more, I won’t say traditional but from a schedule perspective you kind of at least know of what to expect at that point. And then if people are travelling or whatnot, again they can come in the morning for whatever reason they are a little bit late. That component of it is taken care of. And then the other thing that people really really, everybody talked about was they’d like more hands on and we did talk about a hands on. Well, nobody criticize I guess I could see that being one thing that we kind of tweaked a little bit from our original session. The hands on, that lab idea sounds really nice but it is a lot of work. It is super hard to pull off well because you’ve been in the lab, some people breathes through it, others have no idea, some are in the middle. How do you work with all of that?

Steve: And do people bring their own computers or do we supply computers?

Carlos: Yeah, exactly, again if we are supplying computers and all of a sudden it’s much more expensive, and logistics and things.

Steve: And if everybody brings their own then we got to figure out how we get the right software installed.

Carlos: Yeah, exactly or are we installing software things like that, all those kinds of questions. So one of the thoughts that we have and we’re looking for the right people is to that might be a way to partner with our sponsors and to have a couple of things that we say ok in the conference you’re going to be able to come away with. Let’s just take as an example, you’re going to build an availability group, create a container in SQL Server, create an SSIS package and use some DMVs or do some database monitoring. You got to get hands on on those four areas and then have the sponsors come in and help support those things and again I’m just using an example because I’m talking to them. Like the folks at [term unclear – 40:41] Having them come out, they can help with the lab in setting up the how to, this is how I setup my container in SQL Server. They will make that commitment but then they will also be able to then say, ok you’ve seen how that’s happened. Now here’s how we can help make your life easier.

Steve: Yup. You know what I think that would be incredibly be valuable because so often when you’re at the conference and you meet with a sponsor or a vendor and you talk to them and you think, “Wow, you’re product or your stuff seems really cool.” But then you get home then you realized, “Oh, how do I try that out?” or “Where do I go now that I want to try it out”, and you realized of there is not trial to download or I got to talk to a marketing person or something like that. But if you could just have like hands on demo with more than just here’s a lecture from a vendor. Like, here’s how we actually make it work hands on, that would be awesome.

Carlos: But in conjunction also with, ok, so here’s how you could do like SSIS. Let’s build an SSIS package and some basic things. Now here is how you could make that easier. And those conversations and if the sponsors are willing to commit to come to the conference and spend the time to make those relationships then I think it will be easier to have those conversations when it comes for lab time towards the end of the conference.

Steve: So we should be thinking about what sponsors would be interested in doing that.

Carlos: That’s right, and so companeros if you have ideas let us know. And so this kind of gets us to the question of should we try this again in 2018? I can say that I had a blast, right? I enjoyed it. I enjoyed getting to know Erin and Gretchen and Bryce and Jeff and Dave. Good thing we gave him a companero shoutout the other day. Getting to know those guys and of course the speakers in a different way. But ultimately I think we want to see if this has power. Is this something of interest? We’d like to hear from you. I think you can do that in a couple of ways, right? Obviously, social media is one. The other is we’ll make available on the show notes page. We’ll put in a little section. I don’t if poll is the right word. But we will give you the ability to sign up, to put your email address in and say, “Yes, I’m interested in being part of the conversation about the 2018 conference.”

Steve: Yup, and we will not add your name to any kind of spamming list. We will just use that for discussion of where we want to go, or what we want to do, those kind of things.

Carlos: That’s right, and many of you have trusted with your email address and we hope that you’re not seeing that as, you know, we are providing good content and obviously there is a subscribe button there and everything which you don’t want to. But I think if we can get to, and I’m not sure what the right number is. But I think if we can get a certain number of people to sign up to say, “Yes, I’m interested in providing feedback at a minimum.” And would be interested in being contacted or being informed about when and if we do this in 2018; if we can get a certain number that I think that will be one of the criteria that we’ll use to determine if we should go forward.

Steve: Yup, I like it.

Carlos: Yeah. So I think that’s the retrospective. I think we learned a lot. I think I’m glad that we went through it and thanks obviously to those who attended and to the speakers. I guess I should say in that number, so we talked about those numbers, those expenses. The only thing I provided for the speakers was the ability to attend, obviously their food, and the cruise, a t-shirt and a notebook. Randolph came in from Canada, just all these flights, hotels, all of it was on their own. So we can’t say enough thank you to them. If we had those additional cost obviously it just wouldn’t have happened.

Steve: Yup, so what are we doing next?

Carlos: Yes. I guess with that we’re testing out this idea of the SQL Trail. We mentioned this event to talk about at the end. So Steve and I thought, again kind of in that idea of engaging with you companeros and continuing the conversation, we are putting together an event at Summit. So this event is going to take place on Wednesday evening. So we’re actually going to bundle it up. It’s going to start in the middle of the last session of the day on Wednesday. I don’t know we go to 7PM. We’ll put out the details, again, this idea of SQL Trail. I was thinking about, you know some SQL Trail should that be something that kind of call our defense. This will be included on that. We’re going to provide some food, it will be at the, which I also can’t remember the name. But we will provide advertisers and things and we’ll have raffles for drinks and whatnot which we will explain later. But I was thinking, you know what, SQL Trail and then I was thinking about food and my wife actually said you should call it Trail Mix.

Steve: Ah, so it’s the Trail Mix event.

Carlos: There you go, that’s right. It’s the SQL Trail Mix event. So that’s what we’re going to call it and we’ll hope you’ll come.

Steve: And how can people find out more about it?

Carlos: So it will be on the shownotes. If you’re on our mailing list you’ll get an announcement about that. And then through social media we will be publicizing that as it comes around. We’ll invite you to sign up for Eventbrite so that we know approximately how many to expect and again all the other details will be there as well.

Steve: But even though it’s through Eventbrite there’s no admission fee or anything like that. It’s just trying to get people the RSVP that way.

Carlos: Yeah, exactly so just we know and ultimately it will help us with the raffle for example because we’re going to raffle up some of the SQL companero shirts. And the notebooks and things and so that would just the easy way we have your name. We will preprint them and then we can just pull it out of a hat and go from there.

Steve: Yup, awesome.

Carlos: Ok. Well, I guess that’s going to do it for today’s episode. If you have other questions obviously reach out to us. We would love to continue to talk with you. Our music for SQL Server in the News is by Mansardian used under Creative Comments. You can always reach out to us on social media or connect with us on LinkedIn. I’m @carloslchacon.

Steve: Or you can connect with me on Linked @stevestedman and we’ll see you on the SQL Trail or at PASS Summit at SQL Trail Mix.

Episode 114: How do you start consulting?

One of the advantages to a small conference is the ability to take attendee feedback and put it in place during the conference.  We actually made time for the attendees to pick topics they wanted to discuss and this episodes comes from one of our attendees Aaron Hayes.

How do I start consulting?It can be very tempting to think of the good life of consulting.  I almost liken it to playing the lottery–what am I going to do with all that time and money?  While the odds on successful consulting are a bit higher than the lottery, just saying you are a consultant won’t automatically bring in the clients.

The reasons data professionals get into consulting are varied and in this episode we are joined by Randolph West and Jonathan Stewart, former podcast guests, to talk about the reasons we started consulting and some of the challenges along with our decision.

From my own personal experience, working for yourself–whether you consider yourself a contractor or a consultant–is very rewarding but demanding work.  There is no one to tell you want to do, but there are very few security nets as well.  One of the most important ideas I can suggest for those who want to own their own business is–forget the technology, how are you going to help other people?  If you are ok with the idea of focusing on others, then there are great opportunities in store.

Do you have thoughts about jumping into technology?  Let us know in the comments below.

 Episode Quotes

“I hate the lost productivity involved in sitting at a desk all day.”

“Becoming a consultant allowed me to… choose my customers, choose what I want to work on.”

“You can do other things but be known for something”

“word of mouth turned out to be a very bad strategy.”

“A consultant is I’m helping you solve a problem, a contractor is getting some job done for x number of hours.”

Listen to Learn

10:26 Challenges in consulting
15:45 Having your own branding or niche as a consultant
19:08 Difference between a contractor and consultant
42:57 How consultants find their particular customers? What are some of the strategies?

Transcription: How do you start consulting?

Carlos: Companeros, welcome to Episode 114. It’s good to have you on the SQL trail again.

Steve: Carlos, Episode 114, sounds like a fun one.

Carlos: Yes, so we are piggybacking a little bit off of a conversation we had at the Companero Conference now that we are on the other side of it and recording sessions afterwards. One of the interesting things that we did at the conference that I’d like to continue and we’ll talk about more of this in our post mortem, but the idea was we wanted to leave some time or some unstructured time where attendees could submit topics that they wanted to talk about. Which we did, and then the attendees voted on the topic, so everybody got to contribute and then we voted. One of the topics that came up was this idea of “How did you get started in consulting?” And so we recorded this, we have it during the conference and then we thought we would make it available to you the podcast.

Steve: And this one was suggested by Aaron Hayes.

Carlos: Yeah, so Aaron of Chicago, shout out to him. Thanks Aaron for coming.

Steve: Speaking of shoutouts do we have other companero shoutouts, Carlos?

Carlos: We do, I guess so keeping with the conference trend so John Szewczak–I think I’m saying that right. I spent two days with him. John, if I didn’t get that last name right, I apologize. This was an interesting event that developed as well in that some of the other conferences that we go to and again the event was two days, so you get to know people. We were heavy on interaction and one of the other sessions that was developed was SSIS, so John who was an attendee had been doing quite a bit in SSIS and as the conversation had jumped up he had responded to a couple of things so we actually had this session. We actually invited Doug who had given a session previously and John up and they led that discussion on SSIS. There were no slides. It literally started with a couple of questions and they just started bringing up a console and just kind of, “Hey, this is what we do in giving thoughts and experiences.” And I thought it was very very cool that contribution can come from anywhere. If you are following the Twitter notice, and we will put it up on the show notes episode for today but you can see a picture of John up there leading that session.

Steve: Well, maybe next time when we do our call for speakers we should open it up to the previous attendees first.

Carlos: That’s right. In fact, he did say afterwards that he would be interested in coming back and speaking, which we are going to make some changes there to that whole process as well so it would be interesting. Another shout out there, way too many to mention, but this week I attended our local high school for career day. And I’ve actually done this for a couple of years now, so I lived in a very blue-collar area. You can take that as you see it. So lot of families, so dual income, and the majority of the folks in our area are not college educated, so that’s kind of how I am looking at that. The high schools they have a lot of trouble from a demographics perspective. We tend to be on the lower end of the socio-economic scale. This idea of kind of being able to break out and, these kids have an uphill road. Anyway, I was there and we’ve been talking and I wanted to get across or extend this idea of the power of the network and why making connections is important, and then connecting with the right people, and there is an investment in that. We’ve talked a little bit about that on this program on the podcast before. And so what I did, it was kind of through the moment. It wasn’t as planned as what I would have liked to but I went ahead and took a selfie with the class. I posted it up on Twitter and said, “Hey, this is what I’m doing.” I put in the #sqlfamily, and then said, “Please reply, like or tweet”, to let people know kind of that power of community and where it goes, and it was super interesting. Obviously lots of comments came in but then just seeing I think the breadth. So while I was at the school, I was at the school for 5 hours and while I was there and then the tweet continue to get likes and whatnot afterwards. It went as far as Brazil, people commenting, which they thought kind of interesting and then all the various states, and so we had several people just reply, some of the people retweeted and then of course liking. And you kind of do some analytics on that tweet and they were able to see, again, that idea of so of my followers, which I think I have around 700. It went way beyond my own network, right? But it was because people knew who I was and I was asking for something that was really easy to do and they could feel like they were helping. Yeah, it was an interesting little experience.

Steve: Ok, really cool, so I guess the thanks goes out to all of your followers who retweeted and all of their followers who come in and retweeted it as well.

Carlos: That’s right, who connected and got in on that. I really appreciate it.

Steve: Interesting experiment there.

Carlos: That’s right. Ok, time for a little SQL Server in the news. There are a couple of more announcements for Ignite that we haven’t quite gotten through but this one I thought was interesting, and we’ll see how it goes. My understanding is that it’s currently in preview, and we’ve talked about Azure data factory on the program before. Although it was really in its infancy and has definitely gained traction a bit now, but they are now allowing to deploy SQL Server integration packages to Azure. There are some parameter around it but the Azure data factory is ultimately what’s going to be powering that and there are some limitations as to what you can connect to and whatnot. But that idea of being able to get SSIS in Azure is picking up some momentum and I think they will continue to work on it and it will be a little bit more usable for those who are more familiar with the GUI.

Steve: So perhaps it’s time for us to do an episode at some point in the future on SSIS in Azure and get someone who knows a lot about that.

Carlos: That’s right, that’s right. Which interestingly enough we are getting ready for summit, and so that is the time that I like to go out and make connections to some of those program managers, and the MVPs and whatnot. And so I will definitely put them on the list of people to reach out to. We should probably go ahead and mention, we don’t have all the details yet, but we could probably go ahead and mention what we’re trying to do at summit.

Steve: Yes. Well, we’re both are going to be there and we’re trying to sort of get an after hours, some kind of a gathering together. Still working through the details on that right now but a companero and podcast listener get together where we can just hang out and chat for a while.

Carlos: That’s right, do kind of a happy hour type of thing. Again, those details are still forthcoming, however, I will give you a little sneak peek and that is it will be to your advantage to leave a review on iTunes. So we are going to work out some incentives. Those of you who’ve already done that, thank you. If you haven’t we invite you to go over through iTunes and leave a comment or review. That will be key to, not key, obviously you could come out. We’ll do something for those who would love feedback.

Steve: So if somebody wants to attend that after hours drinks, whatever we are calling the event, how can they reach out to us?

Carlos: Yes, so obviously through social media, you can hit us up on email as well. So probably I guess through Tweeter or LinkedIn. It’s probably going to be the easiest way. One of those two I’m assuming you’re on that social media. We’ll make the information available and we’ll put it up on our show notes page for today’s episode and future episodes. Today’s show notes episode is going to be at sqldatapartners.com/consulting

Steve: Or at sqldatapartners.com/114 for the episode number.

Carlos: Yeah, and so that would probably be the best way to get that information and we promise for next week’s episode to have that.

Steve: Alright, excellent.

Carlos: So now that I’m thinking about it I should go ahead and plug, so at sqldatapartners.com/podcasts, you do have the ability to join our newsletter and we can make sure that it goes out through those channels as well. Ok, well let’s go ahead and get into the conversation here.

Carlos: We’re actually here at Companero Conference. One of the sessions that came up, one of the interesting things that people wanted to talk about was a little bit of the question of how we got into consulting, so we have Randolph and Jonathan here, and so we are going to have a little brief panel discussion and talk a little bit about it, and then maybe open it up for questions or how we decided to go. First, we will let you give your, the brief story of why you decided to become a consultant.

Randolph: I become a consultant because I hate working full-time. I hate the lost of productivity involved in sitting at a desk all day hearing people have arbitrary conversations that have nothing to do with work and not getting their work done. I hate the idea of a 9:00 to 5:00 or an 8:00 to 5:00 shift where you are expected to be sitting at a desk and looking at a screen or doing stuff when your productivity might be better later and during the day or even at night in my case. I found myself to be far more productive just working from home or finding a quiet space as just as shared workspace that’s in Calgary that I sometimes go to. And I can get more done in an hour of quiet time than I used to be able to do in a full day of work working full time. Plus I get to choose my customers. I get to work as much as I like or want to and I get to travel.

Jonathan: I’ll give a slightly more PC answer than Randolph gave. There was a lot of hate in there.

Randolph: Intense dislike.

Jonathan: Ok, sound guy, you fix it. Let’s see you edit that. I became a consultant because lot of the same reasons that Randolph said too for me. I hate to not be thinking. I know right, I intensely dislike to not be thinking so one of the things that happened a lot of places that you will do a project and a project will be great, will be interesting. You will do a lot of work and stuff too and then it sees lulls. In those lulls sitting at my desk, you know, you can only read the internet for so long and then you start getting bored. And when I got bored, it’s just really bad when I got bored because I don’t like to not be thinking. I’m always thinking, and when I can’t control what I’m thinking about then I begin to think about other things too like, “How much I’m getting paid?” There are all kinds of other things that would happen when I wasn’t thinking. So for me, becoming a consultant allowed me to, as Randolph said, choose my customers, choose what I want to work on, be in a load up as much as I want, be a tech time when I want because I want to do work, I can do work. I do work in the middle of the night because I don’t really sleep a lot, so it let’s me add work to it and lets me choose the things I want to work on. As technology moves forward I could move forward with it and not be stuck based off of a company wants to go forward with a new product or not. I can choose what I want to work on. I can choose if I want to move more toward predictive analytics or if I want to move more towards visualization. It’s my choice so it allowed me to control my career and be happy with it because literally if I don’t like the work that I’m doing that’s my fault because I chose what I want to do. So that’s the short story of how I got into consulting.

Carlos: I told the story a bit before. It’s always been my desire to be an entrepreneur, the idea is very sexy. We are Americans, right, we build stuff. We make our own things.

Carlos: My first company was actually like a sort of computer repair. My partner is Jeff Morrison, called it Mor-Tech. he kind of started it and I wanted to help him and then that kind of fizzled out because, anyway, it just didn’t work. You are always afraid, you are afraid to make that leap particularly in technology. Like, what if somebody asks me something I don’t know so that helped me back for a while. And then I was working for a law firm and I was in a position where I was seeing the contracts of these consultants that were coming in, and then you get to a point where they start coming to you for questions and you’re like, “Hmm…” Admittedly, some of them were in California so a bad basis. I was seeing California rates coming in and I said, “Well, gosh, I’ll do that. Maybe it’s time to make the leap.” The timing just happened to be such that my wife finally is going to give me the green light and so that’s kind of how we started. I think that was, and I looked back at that and while the desire was there I had some semblance of what I wanted to do or kind of seen what was there. I did not have a great footing or foundation of what it meant to be a business owner, and I wished I would have waited, not waited but maybe did things a little bit differently there initially.

Carlos: Ok, so Steve, do you want to give us, how did you get into consulting?

Steve: Well, I got into it a couple of times. The first time I wanted to try being a consultant and this was 12/13 years ago now, maybe. I sort of quit my job, hang out a shingle, started talking to people and my previous employer ended up hiring me. I ended up doing work for them. It was really kind of a weird consulting scenario because I ended up working for them almost exclusively for 1½ year or 2 years.

Carlos: After you had been a full time employee.

Steve: Right, right, and then when that gig came to an end that was when I kind of learn a little bit about making sure that you don’t put all your eggs in one basket which is one client. And I made rules for consulting from there on, and I broke them a few a few time, but then I ended up getting a couple other gigs with other clients and then ended up going full time with one of those clients as an employee which took me out of the consulting ring. And then about 2½ years ago it was when I got back into it. So where I left that job and went full time. I mean, for me, it was really. I mean, just jumping back in and wanting to work with different clients rather than the day job. And I think that, I mean, enjoy being a consultant because it changes the dynamics of what you do every day. I mean today, any day, I usually get up and started early and I work early and oftentimes by 9AM. I’ve completed more work for clients than I ever would have in a full time job by noon, or maybe by the end of the day. And it’s just the nature of consulting that allows you to do that.

Carlos: Right, right. It is interesting that I think a lot of clients or a lot of consultants will start that way. That was my experience as well with a former employer and you get in to it for whatever reason and those rose colored glasses helped you jumped in, and then there comes a point where you’re like, “Holy cow! What did I just do?”

Steve: Oh yeah, absolutely.

Carlos: I think from there let’s go in and I guess talk about, let’s do challenges first, so challenges of consulting. So you mentioned lulls, you know, there’s lulls at work and I think there is definitely lulls in consulting, right? So what are the challenges that you find being a consultant?

Randolph: Scope Creep, so you sign up for a contract that’s going to be three weeks, four weeks, and then suddenly you’re working 6 months. It’s not a bad thing but it might be not the work that you thought you were going to be doing. That’s the good side of challenges that face you. The bad side is where you’re not working for 6 months at all which happened to me two years ago. And that’s why I got more into acting and making films because I had all the spare time but the challenges that I had to pay for it somehow. I found myself compromising; I only do performance tuning to start doing development again because I used to be a developer, so I had to get back into that and I was rusty. So the challenge is that I wasn’t be able to do the work that I wanted to do but at the same time I was keeping my old skills fresh and learning new stuff as well. I was like upside there and that’s kind of how I got involved with a little company called SQL Data Partners. You may have heard of them. Good bunch of guys, the one is a bit odd, then there is Carlos. Hi Steve! I’m kidding, Carlos is the weird one. The challenge is I didn’t work for six months, that was rough. I won’t lie.

Jonathan: Since they’ve already touch on those challenges too, I’ll add another challenge too because we work for ourselves. Another challenge too is something that I didn’t realized until I was doing it is, people don’t like to pay you when they said they are going to pay you. One of the hardest things about making a jump into independent consulting is understanding that for that first month, two, three, depending who your first client is. You may not get paid. You get paid but you get paid down the line.

Carlos: When they want to pay you.

Jonathan: So not just getting paid and being prepared for that and then to understanding how invoicing works and stuffs too. All of my clients are really, they’ve helped me invoice them because there is a lot of things setup invoice, track your time, and stuff like that too. But at the end of the day, you know, been able to go through and invoice somebody and then able to change the invoice in the fly, something happens and stuff too. For me that was a struggle because that wasn’t my strength. Accounting wasn’t really my strength so I actually hired an accountant. Let them do that. So there is a lot of ways to be handed on your business and stuff there too. Some people do it all by themselves and I commend them because that’s a challenge. But there is a help out there so don’t be overwhelmed. You can even contact me. I’m sure Randolph and Carlos as well, if you got questions you want to do it. I see your card, Randolph. You can be Randolph, and I can be … I’ll make a normal $1 bet, we’ll showing our age. But no, the challenge is though is there is other challenge too besides just getting the work and stuff as well, it’s the stuff behind the scenes. The thing is that from taxes and stuff like that and understanding the things outside of your normal day to day task. Those are some of the small things that come being an independent consultant.

Carlos: Yes, as you mentioned tasks so as an employee unless you get to Kevin’s level, right, or until you get to Kevin’s level. I won’t say they are not micromanaging but there is a certain level of expectation, so as database professionals that obvious that you’ve been given is to take care of the database. Maybe they are not giving you every single thing you should do but your tasks is there and assigned. Once you then kind of make the leap, there is nobody giving you that task anymore. The expectations are we want, “make us happy”, and how do you that? The other struggle there is what should your tasks really be? When you first start, you think, you kind of working as an employee like you do anything. So performance tuning, you write a report, you go and talk to whoever about there connecting to a database, right? Lots of the breadths of your assignments is very wide. Now as a consultant it’s slightly different because of that idea of what value are you bringing? Initially I was like, “Well I am a database person you can hire me to do any database problem.” “I don’t have a database problem.” And you’re like, “Wait a second.”

Randolph: Let me find one.

Carlos: Yeah, exactly. I think coming up and actually kind of narrowing the scope a little bit or defining, having a well defined set of what it is that you want to do is very difficult because you feel like all of a sudden, “Oh my world is shrinking a little bit.” But it’s one of those things were you have to go down. I think of Alice, when she takes the potion and shrinks down, you kind of have to shrink. You think about, you could apply this to, when you think for example David Klee. What do you think about?

Randolph: VM guy.

Carlos: VM, VM work, right? When you think about Pinal Dave, what do you think about? Performance, right?

Randolph: One blog post a day for 10 years, that’s what I think of Pinal Dave.

Carlos: You know, when you think even of, so Patrick LeBlanc, or some of these other folks. There is one thing that they are kind of known for. Do they get to do a lot of other things? Sure, but they have that one little niche. Or Jonathan Stuart, what do you think about?

Randolph: Visualization.

Jonathan: My hair. But no, that’s actually the point. That’s one of this I was thinking about too is, and it actually goes falls right align the things. So Randolph was talking about he wants to do performance tuning, so as you figure out what you want to do become known for something. Brand yourself. Your branding is huge. So I chose my hair to make my logo as my branding, but even still visualization is my thing although I do data warehousing and all kind of other stuff as well. You have that niche that gets you in the door and then you can do other things. But be known for something and then you can do other things as well because what happens is that if you’re just a generalist nobody thinks of you specifically to do that one thing when that comes up. If you want to do that one thing, be known for that one thing and then, “Oh hey, we have X, Y, Z as well.” You may not be able to do X, Y, Z but then I can call Kevin, I can call Randolph, I can call Carlos to help me out do those other things. But that gets you in the door. Be known for that one thing, be very good at that one thing. While Klee is known for virtualization he does all kind of other stuff as well. But the virtualization has got him in the door because that’s what he is known for around the world. When you say Klee, you think of VM. VM ware has changed its name, Kevin say as KleeM ware because it’s what it has become. But that gets him in the door, that’s his brand. That’s what he is known for but he is phenomenal with all kind of other stuffs as well. Be known for that one thing, but be able to do other stud as well. Even with Randolph, like you said, he got to do performance tuning but he can do other stuff as well. So if performance tuning gets him in the door and he goes like, “Hey, I want to do X, Y, Z and expand to Azure SQL Database.” He can do all that as well. So once it gets him in the door he could say, “Hey, I had all these other services as well.” But that’s what got him in the door, that’s his brand, that’s what he is known for.

Steve: Right, so I think some of the challenges and I think some of the others mentioned this, things like scope creep. I mean, oftentimes you will start on a project that you think is going to be one thing. And then after meeting with the client and understanding what they want, it ends up growing into two, or three, or four or ten different things that you never even expected. And that could be good or bad depending on how you look at it. I mean, it’s good and it brings in more work but oftentimes it can be bad because one of my rules is never do more than, well hopefully more 30% of your time with any one client. Sometimes I’ll break that and go to 50%. But when the scope creep goes the point that they want you to be there as like all your time with one client, I just don’t go for that because that’s way too risky. Because when the relationship comes to an end with that client which it always does eventually then you’re sort of out of work at that point. I kind of fight that scope creep by pushing back and prioritizing things but then limiting my time amongst multiple clients. Which I guess sort of brings up one of the other challenges that have come across in consulting is, when you’re working part-time with teams that are working full-time, and oftentimes they kind of expect that you are on the same full time schedule they are, even though you are a part-time consultant, and trying to sort of balance between the different clients so that you can responsive as needed but also so you can spend the right amount of time with each client each week.

Guest: What is the difference between a contractor and a consultant?

Randolph: Well, a consultant I feel gets to choose what you want to do, a contractor gets told what to do. You still have billable hours and you still have to fill in a timesheet. It’s all the same thing. It’s just who you are working for is a little but more blurry because a contractor might work for a contracting house like I use to and then I get told, “You’re going here tomorrow. You’re going to work six hours, you’re going to bill eight.” Hey, I’m not going to lie. That’s the difference for me. A consultant is I’m helping you solve a problem, a contractor is getting some job done for x number of hours. The religious work comes in whether contractors are actually doing anything useful, and some are and some aren’t. It’s just depends who you’re with I guess.

Carlos: You think these goes to that idea of like of the branding? This idea of are you being paid for your hands or for your brain in a sense, right? So if you’re being, like brought in, here’s the work we want you to do. I tend to think of that as the contractor route. If it is, “Hey, I have this problem how do you think we should solve it?” Ok now, I’m in putting it into the process and again not a generalization but that’s my delineation. You feel good about that?

Jonathan: That’s exactly how I say it too because some people say that’s semantics is the same way but a contractor is like a staff person. You are filling a role, you are filling spot so they can just throw work at you. As Randolph said, you are being told what to do whereas the consultant is solve a problem, help me solve a problem. You know, like management and consultants, they are all helping solve problems whereas the contractor is somebody who is filling a seat for a specific time and a role.

Randolph: Yeah, there’s nothing wrong with doing contract work if that’s what you enjoy.

Carlos: Yeah, and I guess that’s an interesting point. It’s not like the one is necessarily better than the other. What are you constraints? What are your box? Where do you want to be and then that’s going to make your decision there for you.

[unclear conversations]

Jonathan: I used the same approach in regards to vendors. So we are going to do an imaging project and we sent RFPs out and have them to kind of do demos. I look for things like contractor versus consultant in the vendor aspect. Are they are going to help us solve problems or just put the imaging system in and leave. I think I can use it both ways in regards if you’re interested in consulting like myself. You can still see that in the vendors you deal with. One of the things too that a consultant has to worry about and a contractor doesn’t even though they can actually be the same person at the same time as well but the consultant has to work on managing relationships as well, if that’s at the end of the day you don’t want as Douglas is saying, you don’t want to just come in and fix something because that’s the end of the gig there. I want to develop a relationship so I would rather spent four hours with you every month than spend five hours with you one time upfront just to do one thing. That’s another big thing there too so.

Steve: I think that, I mean, as a consultant you are coming in to do something that sort of project or problem base where you are trying to come in and create something or fix something or come up with a solution, whereas, a contractor you are punching time on the time clock. And not that you’re not getting things done but you’re showing up and you’re there on the clock for whatever is needed. And I think the comment was, are you paid for your hands or your brain? And I think that that was a great analogy there in that, if you’re there just sort of on assignment for a certain amount of time and you will do whatever needed in that time, that’s kind of the contractor role. And when you’re there to help get the job done to focus on a specific task then I think that’s more on a consultant where you’re either building something or advising on something or coming up with solutions, and that consultant role is the piece I really like.

Carlos: Great, and again, so if you we did mention it, I should mention it here that neither is right nor wrong. The labels are not really important per se. I think it is however more of a philosophy of who’s kind of driving the boat in the relationship in the sense of who’s dictating what work will be done, and that can be a fine line sometimes. I think kind of going back to your comment about the expectations of the client dictates quite a bit about of what role you’re taking on.

Steve: Right. I think some of that comes from just what the client has been used to working with. I mean, if they’ve been used to working with a staff type position where they get a body and can work specific hours. That can be very different than getting in a consultant who’s there with a specific specialty like we have to solve a problem or fix something or help them with their performance whoever it may be.

Kevin: Kevin Feasel, professional podcast guest. You guys just talk about… it’s better than being an amateur. So you guys just talked about brand, about building up skillset but I would like to bring in the next question as somebody who has absolutely no interest in doing consulting work, hello employers, how do you go about finding the particular customers that you want to work with? How much of that is them finding you, how much of it is you finding them? What’s the strategy behind that?

Randolph: Most of my customers were from LinkedIn which surprised even me. I thought I’d find people through word of mouth, those were terrible customers because what happened is somebody says, “Oh, you know computers, I know somebody who could use your help.” And then you end up working on MySQL doing index tuning and then after 20 minutes you got no more work for them because… They are only useful in certain context. But the point is that word of mouth turned out to be a very bad strategy. LinkedIn where I specified this is what I can do, this is what my company does. People who need it usually go there because they are business peoples so they are looking for a problem solver for that particular problem. I do give my cards out that’s why I have them up here because if anybody hasn’t got one yet I’m going to give it out because I want people to associate me with the fact that I am available to do almost anything. Because the chances are 5% of the people call me back and out of those 1% or 2% of those 20% of those people will actually turn into a paying customer. So I have to go and do the whole mail shot or whatever so that I can get 1% back to actually pay me to do something. There is a lot of targeted advertising I could do but I don’t have the time or the money to do that right now, and I don’t want to because I don’t want to work full time.

Jonathan: Because if you work full time he will never be able to play dead in TV shows and movies. It’s acquired arts skill. Don’t you play dead with me sir. There is a wide gamut of ways that you can acquire customers. Most of my customers I do partnerships with a lot of people. A lot of people find me because I speak a lot so I go places and stuff too and people say, “Well, come up to …”, and I talk to them too. A lot of these are business owners. They have their own SQL practices as well. But it’s like, “Oh hey, we might want to partner with you. We want you to do business intelligence. We may want to do visualization. Can you help me with this client and do such and such.” Speaking is a great way to get your name out there, to be known for certain things. So for me, like I said, a lot of my clients come through not just word of mouth but from partnerships that people I’ve met on the speaking trail, or heard me on SQL Data Partners Podcast, right, I got people contacted me in that way as well. It’s not really hard to find me because I’m the guy with hair. See it’s a branding thing again, right, but have a brand. Everybody has a way to be branded you just haven’t thought about it yet. So if you really consider doing this, think of a way to brand yourself to get out there and then there’s multiple ways. As Randolph said, he gives his card to everybody. Anybody can give us cards because that’s a cheap way to advertise. You know, me, it’s not really a cheap way because I’m flying everywhere to do this. But there’s so many different ways. There is not just one way. You got to think outside the box. You know, and there’s more ways than we talk about. I’m sure Carlos has another way as well.

Kevin: Write a book.

Carlos: Initially we tend to suffer. It’s very similar to the idea of finding a job. How many of us has just taken a job that has come our way? Do we actually think about the industries that we want to work in? They type of people that we want to work with? The type of projects that we want to take? We might say that, “Oh yeah, I want to do migration”, so SQL Server 2017. At the conference several people have mentioned, “Well, they approach me with this job. I wasn’t really looking for it but it paid me x percent more than I was making”, and so you take it. We follow on that same mindset that we want to spread the word, the cards are great but ultimately it is trying to, with that niche, finding what you want to do, we’ve had our best success then defining a vertical which is healthcare and then going after those people, right? And so, yes we still like I still hold the conference, I still do the podcast. Our health care people, in fact I don’t think if there is any health care people here?”, so we’re still doing things outside of healthcare before my marketing and like where we’re going to look for customers. We spent all of our focus. In fact, we’ve actually even gone as specific to customers of GE. And that’s who we’re targeting, trying to establish those relationships, increase our network width. We want to become the SQL Server people for GE centricity customers, and then once we do that we will go on to the next phase.

Steve: I think finding customers, that’s the hardest in the very beginning because you start out, you have no history there other than with your full time employer before that perhaps, and you’re trying to find customers. I think that getting the on boarding process for new customer the very first time always seems to take far more time than you ever expect because that first customer is everything to you, and without that first customer you have no income, you have nothing there. And then I found overtime once you’ve got that first customer, as long as you’re not breaking the rule of putting all of your eggs in one basket with just a single customer, the next customer is little bit easier because you have the work from the first customer to sort of subsidize the time while you’re waiting for that second customer to get things signed. And then once you’ve got more customers you end up getting repeat work. And I think there are some customers that you may work with that once you found them you’re getting work every single week for years to come, and there are others that once you found them and you’ve done work with them it may be you come back and work with them every six or eight months depending on what their needs are. I think that finding them, and I think a lot of it is really just the specializing. And I think in finding the customers you just find out and say, “Hey, I’m a computer tech guy. Hire me.” That is I think the wrong approach there because then you’re competing with everybody that’s a computer tech person anywhere. But if you specialize, like my specialization has been in database corruption. There is not a lot of people to compete within that area so that’s an area that people find me now when they come across corruption. And because of that specialty I’m able to take it on and fix them and get happy customers really quick right there or not always real quick, but generally as quick as we can. The other areas like the database health assessment. A lot of that has come out of Database Health Monitor. I think that just getting your name out there, whether it’s like what we do with the podcasts, or what we’ve done with the database health monitor or blogging. I mean just getting out there, social media, LinkedIn. Things like that to get your name out as someone who very rarely when you’re doing those kind things turn into an immediate like you do something and within 24 hours you have work. But it turns into you become that trusted name and then 6 months or two years later somebody remembers, “Hey, that’s the person who specializes in that. Let’s call him.”

Carlos: It’s definitely an investment, right? You have to wait a long time for those returns sometimes. It can be very very difficult. Ok. Well, thanks again Aaron for the question. I thought some interesting insights and conversation if you’re interested in talking a bit about this more. Maybe have some of questions, of course we much being at the summit or you can reach out to us via any of our social media channels and we’ll be happy to have a conversation with you. Give some thoughts around that.

Steve: So one of the things that I wanted to do was call out an interesting comment that I like from Randolph where he was describing us at SQL Data Partners and he says, “One is a bit odd, then there’s Carlos.” Thanks Randolph, I love it.

Carlos: Special thanks to Randolph and Jonathan for providing some insights there as well. Our music for SQL Server in the News is by Mansardian News under Creative Comments. If you have other suggestions for topics or suggestions about things we should be talking about, you can reach out to us on social media. You can find me at LinkedIn, I’m Carlos L Chacon.

Steve: Or you can find me on LinkedIn as Steve Stedman, and we’ll see you on the SQL Trail.

Episode 113: Standing on the shoulders of giants

“If I have seen further, it is by standing on the shoulders of giants.”  Those words are attributed to Issac Newton and they are fitting for the way I was feeling recently as I thought back to those who have helped me in my career.  While we might always want things to be better, as technologies we have it pretty good.  There are unnumbered people who have dedicated untold amounts of time so we can have the tech available to us.  We take a moment to think about those who have paved the way for us to be where we are now.

Whose shoulders are you standing on?  Let us know in the comments below.

 Episode Quote

“This idea that the way he was able to look at the world is based on all the previous work that has been done”

“I guess “gratitude” is the right word of those who have gone before us and enable us to have those who are in technology we have a pretty nice career.”

“There are just a lot of people out there who just want to help and just want to better things with what they do every day.”

Listen to Learn

00:38 Companero shout out to David Stoke
02:11 SQL Server in the News – SQL Server 2017 is out
02:35 Interleave execution for multi-statement table value functions
05:34 Artificial intelligence, analytics, machine learning
07:58 Today’s Topic: Standing on the Shoulders of Giants – having that gratitude to the people who have gone before us
11:27 Carlos shares a documentary he watched about bitcoin
14:50 Stop complaining
19:56 Who are “giants” in the life of Carlos?

Transcription: How do you start consulting?

Carlos: Companeros, welcome to Episode 113. It’s good to have you on the SQL trail again.

Steve: Yes, it’s good to have everyone who’s listening. Good to see too, Carlos.

Carlos: Yes, always good to connect back with you, Steve. Today, we’re going to be talking about, kind of changing the subject up a little bit, and the idea is standing on the shoulders of giants. We’ll get it to more what we mean about that in a bit later but that’s the idea of our topic for today.

Steve: Ok, sounds good.

Carlos: So we do have a companero shout out. I want to give a shout out to David Stokes. Now as we record this we actually have not had the Companero Conference just yet. That will happen, actually as soon as I hang up I’m going head down there and have the conference. But I want to give a shout out to David. David is the leader of the Norfolk user group. He was actually the first person to buy a companero ticket and has been a great supporter of the podcast and of the conference. I want to give a shout out to him and thank him for what he’s done.

Steve: Yup, you know the interesting thing about David is that he is the very first user of Database Health Monitor that I have ever met in person.

Carlos: Oh ok, interesting

Steve: I met him at Charlotte, North Carolina for PASS Summit, was that 3 or 4 years ago?

Carlos: That’s right, 2013 I believe

Steve: 2013, ok, so 4 years ago. Yeah, it was really kind of cool to meet him and listen to what he had to say about Database Health Monitor back then and he’s been using it ever since. David is a friend of Database Health Monitor.

Carlos: Yes, very nice. Ok, as mentioned, I’m sure the conference went great. You already have that and we’re looking back to some feedback. We will probably have some kind of post mortem about that. But now I think it’s time for a little SQL Server in the News. We’ve been going back and forth a little bit about this and one of the things I wanted to talk with, of course we know that SQL Server 2017 is out and should be up, right? I’m sure you spent your weekend installing that. But one of the things that I wanted to discuss a little bit here which has been mentioned I think it’s expanded a little bit here and that is the interleaved execution for multi-statement table valued functions.

Steve: Oh yeah, this sounds pretty cool because there’s been so many concerns over the behavior of multi-statement table valued functions over time.

Carlos: That’s right. Yeah, so ultimately the bad news is that in previous versions of SQL Server, and I guess in 2017 as well. I mean, there is a fix for it but the optimizer can’t always determine how many rows the table valued function is going to affect. As a hard time just in for that fact, right, and so in previous versions basically it would guess a hundred and then it would really be 5,000 and so we have a bad plan, and that was kind of the issue. So what they are going to do here with 2017, they are adding some analytics and some machine learning basically into the optimizer, so that when the first time the plan comes in with the table valued function it’s going to guess again a hundred rows when it executes it’s going to have 5,000 rows. The difference now is that the second time it runs it’s going to take a peek at that previous plan and say, “Hey, how did I do?” And it says, “Oh, I did horrible. Let me adjust my numbers and take in the previous execution into account to see if I can come up with something better.”

Steve: You know, I wonder as I hear that if that is actually going to help or hurt because. I mean, if you table valued functions commonly have the same number of rows they are processing it might help you. But what happens if you’ve got such a wide variety of data in your database that every time that gets called the table valued function has a very different set of data that it’s processing depending on a client or customer or whatever it may be.

Carlos: Right. Yeah, and so obviously we have skewed data it makes it difficult to solve for the outliers. There still probably an outlier case. I think here what they are trying to get at basically is do I use a hash or do I use nested loops? Which way that I’m going to go about that? I think, again, after you execute it a hundred times you’re going to have some data there to then be able to say, “Ok, well, 80/20 rule. This is the way I’m going to go with.” I guess the implications here are not suggesting all of a sudden that everyone start using table valued functions and this is going to solve everyone’s problems there. I guess what was curious, or what was interesting to me is that I thought this was a very interesting way because we hear a lot about artificial intelligence, analytics, machine learning even. This is now a problem that we deal with all the time and now Microsoft is putting that analytics, that machine learning into the product that we love. And I thought, you know what, I meant that’s kind of a very specific scenario. I wonder if we couldn’t start taking some of the same thought and then applying it to other areas of our business so that our users or our customers might also benefit, so that was kind of my thinking there

Steve: Right, so it will really be interesting to see if somebody comes out with like an AI package for tuning SQL.

Carlos: Yeah, something like that, right. Or even that idea of, “Ok, I have this data. My customers keep asking me for whatever use cases. Can I apply some analytics here to give that better information? So it will be interesting to see how this continues to evolve. I think this idea is going to affect those of us who continue to be administrators or gatekeepers of data. These are going to be problems that we’re going to have to start solving, so it will be interesting to see how other tools crop up and try to address other problems.

Steve: Yup. Yeah, and I think it’s been easier with just sort of fix rule sets previously to be able to definitively say this is how something behaves. But with AI built in on how those things are being handled and how it’s tuning itself almost, it may be harder or maybe more work to figure out exactly what’s going on when there is a problem

Carlos: Sure, and I think that’s goes back to you are still going to have to know your system for those outlier cases. What’s going on? Why are you doing what you’re doing and is there a different way to potentially go about it?

Steve: Yup, interesting.

Carlos: Ok, so today’s episode, the show notes are available at sqldatapartners.com/giant

Steve: Or at sqldatapartners.com/113 for our episode number.

Carlos: Yeah, so ultimately this idea, and maybe a became a bit nostalgic which we can get into in a second but Sir Isaac Newton is quoted in 1675 as saying, “If I have seen further it is by standing on the shoulders of giants.” And ultimately this idea that the way he was able to look at the world is based on all the previous work that has been done, right? Helping him to arrive to where he is now. I got a bit, as I mentioned, nostalgic about this idea where I ran into to somebody who work with a former co-worker of mine. We started talking, and so this former co-worker is actually the person who convinced me to apply for my first database job.

Steve: Oh, I remember that story from one of our previous podcast.

Carlos: Yes, and I wasn’t actually completely qualified for it. But he convinced me and he coached me and said, “Hey, I think you do great.” And I think back to that, obviously, you know, there are thousand different I could have gone but I think about where I am now and ultimately kind of that first step of making that decision. And so I think about it, I guess “gratitude” is the right word of those who have gone before us and enable us to have those who are in technology we have a pretty nice career. Yes, it’s not all roses and sunshine all the time but the fact that those giants have gone before us, many of whom we’ll probably never know. We have to look up in Wikipedia, and the fact, even the first people who put together the rules for the transactional database. I know he was brought up in a keynote, unfortunately I can’t remember his name, obviously on top of my head. But, I mean, they put in place what we have today. And I think it’s easy, and not to discount. I don’t mean to discount them at all but it’s easy to look at the Bill Gates, and the Steve Jobs, and the, what is it? Oh gosh, Linux is going to kill me now

Steve: Linus Trovalds.

Carlos: There you go, thank you.

Steve: I don’t know if I pronounced that right

Carlos: If not, I’m sure you’ll get corrected. Again, not to discount their contributions but they are not the only ones. You think about all the teams behind them that helped them do what they wanted to do.

Steve: Yeah, I guess with that, I mean you look at a product like SQL Server, it’s really been like a Microsoft product since 1989/1990 but it was being worked on for years before that and then before anybody even started working on what became SQL Server, there were people in the 70’s that were theorizing on how these databases would work and the rules around it. I mean, 50 years worth of work, and foundation that has gone into what we have today in SQL Server 2017

Carlos: Right, and how amazing it is. It was interesting to kind of [term unclear – 11:26] here slightly. I was watching a documentary on bitcoin, and so bitcoin is the digital currency that they are trying to advance and I didn’t realize but, oh gosh, this Swedish guy who started Wikileaks. Whatever his name was, he was part of the original group of this eight people who put together sort of theorizing about digital currency, and they came up with a block chain idea and methodology. It was because of his situation and what Wikileaks was doing that really kind of put bitcoin on the map because when PayPal took their services away and said, “Hey we’re not going to let you use PayPal services.” That’s when he was like, “Ok, well, bitcoin, Internet can you help me,” and kind of launch that whole idea. In a way it was very interesting. But anyway, in the documentary they go and they talk about and they say, “The person through the door with a new idea get shot.” Anytime you have something revolutionary, the first person through get shot but somebody has to go through the door. So they talked about some of these first people like the first companies that started in New York City trying to exchange currency. For example, dollars to bitcoin and things like that. Well, he’s going to jail. Long story, but you can watch the documentary. Anyway, but I thought it was interesting. I guess going back to this idea that all of the ground work that was put in to enable us to enjoy what it is that we enjoyed today.

Steve: Right. But what was interesting, I mean, he was going to jail for Wikileaks side of things not for bitcoin side of it.

Carlos: Yeah, I’m sorry, so Julian Assange. That’s his name, the Wikileaks guy, and he is going to jail for something else. I don’t know it. But the guy in New York City who was doing the bitcoin exchange, so he was actually trying to get people on the bitcoin and basically sell them, bitcoins this idea. He is going to jail because one of the guys, unfortunately, illegal activity tends to push a lot of innovation. And so when you have something that is not monitored there was some illegal activity going on. He knew about it, and he was basically selling bitcoin to people who were doing illegal things and so that’s why he’s going into jail.

Steve: Oh, interesting. Ok

Carlos: It wasn’t that he was doing anything illegal but he was kind of aiding, and abetting if you will in a sense.

Steve: Wow! I hope I’m not considered aiding and abetting by investing in bitcoin. Or maybe the word is “gambling” not “investing”

Carlos: Yeah, that’s right. It is still a bit shaky, again to progress for a minute, the value of how that gone up it was pretty impressive but obviously it has come down I think in 2011 or 2013 something like that. And so where I was going a bit with this as well is this idea of it kind of grits on my nerves a bit, and we all like to complain. We all find ourselves in situations we don’t want to be. When people will say, “Gosh, I they had only done this right. Why they have to do it this way?” Basically, I’m inconvenience and I don’t really appreciate all the things that I have

Steve: Right, and sometimes that inconvenience comes from just not understanding what the people did before you and just complaining about it. I mean, I’ve done that but I’ve seen that happen plenty of times.

Carlos: Yeah, that’s right. You have that “Aha” moment where you’re like, “Oh, so that’s how it works.” Sure, you may have developed it differently; you may have architected it differently all of that jazz, great. But, you know, you work there, right? And now you’re now kind of taking advantage or taking over in some case. No I’m not saying it’s all always great but the decisions that were made there before you have you put you in a position where you can now either take over or ride that bus if you will. Yeah, it’s kind of interesting.

Steve: Yup, and you may look at something and say, “Wow, that’s stupid. Why would they have ever done that?” But then you look back 10 years ago at what was being done, or 15 years ago, and that was cutting edge technology and the fact that it even worked at all was completely brilliant.

Carlos: Exactly, exactly. Yeah, you know it is funny particularly as database people, like yourself Steve, there are developers among us or recovering developers or whatever you want to call them. But we like to complain quite a bit about you know, and the stupid developer if you’d only done that. Well, I can kind of find that we’re being a bit of a hypocrite because we use software all the time, on our phones and on our computers. I mean, we wouldn’t be able to do anything that we do now if it weren’t for software. We need to be mindful of where we are, what has been invested and give people a little bit of break when they are not upgrading to the latest thing. I guess in our last episode even was about patching. You know, just staying up and keeping with the latest and can kind of get a little tiring sometimes, and so you’ll forgive people a little bit if they are not quite there to where you are.

Steve: I think some of that comes back to what I learned in a class. It was called the Dale Carnegie class that I took several years ago. It was based on “How to Win Friends and Influence People” book as well as some other books, and it was a simple statement that said, “Don’t criticize, condemn, or complain.” It’s just that one of those things that just comes up and if I find myself criticizing, condemning or complaining, I try and stop myself. I try, so it doesn’t always happen but I try.

Carlos: Yeah, it’s one of those things, the reality in today’s world with social media everybody has a voice box or a [term unclear – 18:11] and they want to get up. It’s not to say that reform is not needed, that we can’t make progress. There are times when change needs to happen. There’s no question on my mind. I think it is just more about the way that we go about it and then recognizing what we have and how do we go from there.

Steve: Yup, and you know, no matter how bad the day may be we’ve got things pretty good right now. I think that’s so often overlooked in, well you mentioned social media, I mean, if you’re on social media and it’s almost like everyone is out there just to complain

Carlos: They are either having the best day of their life at the beach with the vacation photos or the world is against me. I think those for you companeros, those who are listening, going to be in technology, yes, you may not be where you want to be. And I’m not saying that you should settle. That’s not what we are saying either but I do thing that you if you take a minute to use the phrase from our culture if you will, is to capture many blessings and just to kind of appreciate where you are and know that there is a path ahead. If you will continue to work there, like we talk about in the beginning, all these artificial intelligence and analytics; yes, is it going to affect you, is it going to change what you do? Absolutely, but hopefully that will be to your advantage and not to your decrement.

Steve: Right, yup

Carlos: Anyway, we thought we change us up just a little bit and share some of our thoughts on that topic. I guess we’d like to, again, thanks to all those shoulders we stand on, so Robert Pollard, he was the guy who commits me to get to that job interview. I even think about Matan Yungman and Guy Glanster. The other guy is the SQL Server Radio. They are out there doing the podcast. Listening to them made me think, yeah, let me try this. Right, let me do this as well. And then of course all those in the community who have give freely of their time. I guess especially thinking about that to all of the speakers who came to the Companero Conference. They put a lot of time and energy and I’m super appreciative to them and what they’ve given.

Steve: And you know, it’s just interesting because one of the reasons that I enjoy what I do so much I think a big part of that is because of the SQL Community. You don’t see, at least I have never seen a tech community similar to what we have with the SQL Community. There are just a lot of people out there who just want to help and just want to better things with what they do every day. And I think that’s awesome

Carlos: Yeah, that’s one of those things. I think standing on the shoulders, again, I’m sure there will be a lot of people out but I kind of, if there’s one person I would point to, Steve Jones. So Steve Jones, probably one of the most humble guys you’ll ever know. He has done quite a bit in the community, kind of a very public figure but he is always looking to help and do so in a way that he’s not boastful. He is just trying to help you along and he has taken his breathe of experience and freely gives of that and all things to do this.

Steve: Yup, and what’s interesting with Steve Jones is that when he presents and gives back and shares with people. He does a great job in the tech side but he also does a great job on like the personal improvement side. Like, how do you get out there and improve the position you’re in – how do you improve your social media footprint, how do you improve your value, all those kind of things. I’ve seen him present a few times and definitely another giant to call out there.

Carlos: Yeah, it was funny, so we have him on the podcast that was one of those first ten episodes and looking to expand your game. And one of the things that I really liked about Steve, because again, he is one of these internet famous folks and he is also one of the few of that genre that will come up to you and say, “Hey, I’m Steve, what’s your name?” You’re like, “Ahhh. I forgot, I know who you are.” Of course he is in a Hawaiian shirt. He has a great brand but part of that brand I feel like is his contribution, and because of that he has made it the norm or almost the expectation of this is what the community does.

Steve: Right, yup.

Carlos: So awesome, so that’s our episode for today. Companeros, thanks as always for tuning in. Our music for SQL Server in the News is by Mansardian used under creative comments. If there is other topics you would like for us to talk about or ideas, you can hit us up on social media. I am carloslchacon.

Steve: You can connect with me on LinkedIn stevestedman, and we’ll see you on the SQL trail.

Episode 112: Keeping up with patching

When listener Mel Vargas first suggested the topic of patching, I was not sure this would work as a topic; however, the Equifax story had just developed and there are many other security related issues that could be prevented with patching.  Match that with our guest, Robert Davis, who just happened to publish some articles on patching and I decided we needed to do this episode.

I should never have doubted this as topic.  While we are a bit more protected in SQL Server than others, the security threat is still real and this is something everyone has to go through.  Robert presents us with some interesting details on how he goes about patching and we think you will find the episode compelling.

Episode Quotes

“You not being able to be patched for the new stuff, you are really opening yourself up to a variety of attacks.”

“We’re all on the same team ultimately and the success of the application really benefits all of us.”

“There isn’t a security reason to be diligent about patching SSMS.”

“The really good DBA, really proactive one, gets overlooked because everybody else in the company doesn’t see the fires that you’re preventing.”

“Learning to speak up for myself and to publicize the things I do definitely was one of the best things I had to learn how to do.”

Listen to Learn

01:08   Episode Topic: Patching
02:19   Why should we care about patching and what it helps to do?
06:26   Maintenance window, automated patching and system center
07:56   Octopus: Patching Automation Tool, automation tools
2:52   Deploying and rebooting, and restarting services
15:48   Do you really need to update SQL Server Management Studio for security reason?
19:44   Cumulative updates in Azure
22:34   Windows patches and SQL Server patches, failover cluster, availability groups
26:39   Patching testing
29:07   Scenarios and issues when trying to do install updates or patching
36:32   SQL Family questions

About Robert Davis

Robert is a SQL Server Certified Master, MVP, and has spent 17+ years honing his skills in security, performance tuning, SQL development, high availability, and disaster recovery. He served as PM for the SQL Server Certified Master Program at Microsoft Learning, and in various roles at Microsoft specializing in SQL Server administration, development, and architecture. He currently works as a Database Engineer at BlueMountain Capital Management where he spends a vast majority of his time tuning massively parallel queries. Robert feeds his passion for security by acting as co-leader of the PASS Security Virtual Chapter.

Episode 112: Keeping up with patching

Carlos: This is Carlos Chacon.

Steve: I am Steve Stedman.

Robert: I am Robert Davis.

Carlos: Robert, welcome to the program.

Robert: Thank you.

Carlos: Yeah, it’s good to have you. Thanks for taking a little time of your busy schedule all the way from New York City to come and chat with us today, so thanks for stopping by.

Robert: Thanks for having me, coming at you from the heart of Manhattan.

Carlos: Wow. Yes, deep downtown.

Steve: Well, I know over the years on different topics that I’ve googled on around SQL Server often times some article or post that you’ve written has come up and definitely appreciate that, it’s always been good content. I want to say thank you.

Robert: Wow, thank you for that feedback.

Carlos: Yeah, and it’s nice to have you on the program today. Ultimately our topic is going to revolve around patching. Maybe not the most sexy of topics but as administrators something that we all have to live with, and I think has only accelerated with the “new” Microsoft and not only we’re getting patches now, we’re actually getting versions and all these things. And so keeping up with all of these can be laborious at times. And so we wanted to start this conversation with talking a little bit about patching. And maybe let’s just start with the basics, right? You know, there’s an inevitable or at least in the past it has been I’m going to wait to upgrade until I get SP1, right? And then, we’ll go forward with the upgrade. And so I think a lot of people look at patching for lots of different reasons. I guess let’s just go ahead and hit on a couple of ideas of why should care about patching and what it helps to do.

Robert: I’d like to just point out the fact that we had a couple of major events very recently in the computing world where there were vast instances of people’s servers being encrypted and it was utilizing very very old vulnerabilities that have been patched years ago that the people affected by them were the people that weren’t patching their servers.

Carlos: Or even learning older versions of older operating systems and things like that and then just have them patched.

Robert: Right, exactly. Those were two major events that could have been stopped right away if people just have their systems patched up to date.

Steve: Interesting. So I mean I have come across situations where people have said, “Oh we’ve got SQL Server 2005”, running on whatever older version of Windows Server it is that they are happen to be running on, “It does everything we need for our business and we don’t need to spend money on patching or updating it because it’s getting the job done for us.” What would you day to someone with the comment like that.

Robert: I would say to those people that if for these really old versions that are no longer supported that means you’re not getting patches for these SQL instances anymore. When you see things like these encryption attacks or SQL-Injection attacks and a lot of these know vulnerabilities attacking people’s systems. In fact, just recently we had a reoccurrence of a mass Slammer attack. If you remember, Slammer hit SQL Server 2000 pretty hard, and suddenly it made a comeback earlier this year. And obviously the people that are going to be affected by that are the people running SQL 2000 that aren’t patched and aren’t getting patches anymore. So by not being on the current version and you not being able to be patched for the new stuff, you are really opening yourself up to a variety of attacks.

Steve: Yup.

Carlos: And so then this kind of gets in to the heart of I think some of our conversations so some of those vulnerabilities. So the worms that were introduced earlier this year there was the patches available 30 days prior. So if you would patch in those 30 days it would have been fine. Some were much older, then we get into that idea of how often or how frequently should we applying them and the testing that goes around with patching? And again I say this is not, I don’t know if anybody that really enjoys patching. I don’t think I’ve met a person that person, right? So I guess thoughts on how can we introduce a process or a culture even to get those patches in and get them tested and maybe what’s necessary to kick the tires, because we know that you as much as we try a lot of times our lower environments aren’t quite the same as our higher environments and that’s a problem for a lot of people.

Robert: Yeah, exactly. But I think really the key to getting people on all our kaitens of regular patching is to introduce automation to the process. Obviously if you’re up late at night installing patches manually waiting for them to finish and then rebooting is going to be a boring tedious task and nobody is going to want to do it. We use system center to push patches out to all of our servers, and with system center we have the ability to. If we have some exception like we don’t want it to be automatically patched and rebooted in the middle of the night we can exclude certain server so we can manually do some certain server but otherwise at a certain point our automation kicks in says, these servers aren’t patched, patch and then reboot them starting at 11PM on Saturday night.

Carlos: There you go, so kind of finding that maintenance window and then making that happen. So you mentioned the system center, right and obviously if you have that product and I’m sure it’s doing a lot more than patching there is some inventory kind of helping you with your server environments just in general. Have you played around with any of the other patching automation tools by chance?

Robert: Yeah, actually when I worked at Microsoft we used Octopus for our patching processes. We would have Octopus, it would push the patch files out to each of the servers and install them and do anything. Octopus is great because if you need to do anything like stop a service before you patch and then restart it after you patch, you can build that right into the Octopus script. So for example, say if we needed to stop a web service that’s getting the web, the database server before we patched and rebooted it we could do that into the Octopus deployment script. It would stop the web services, install the patches and reboot it and then restart the web services on the web servers. It could do the whole, anything that could be automated that could do and the really great thing about Octopus too is we could build into it. Some sort of some basic testing at the end of the deployment like it could go in and send a request to the web service and check to see if it gets a valid response back so that we know. Yeah, we know that the web service. We would setup like some dummy process on the web server that Octopus could hit it would make a database call and it will just return a value, and if it returns the expected value then we know that the database and the web servers are all functioning properly.

Steve: Very interesting. I mean, I have worked in environments were Octopus deploys are used but I’ve never really thought about using it for updates. Kind of makes me want to go back and take a look at it now.
Carlos: Right, well I think it’s a good point too that Robert makes in the sense that you want to automate those but then with automation comes with responsibility in the sense of then going and testing some of those things. I guess there’s a little bit of that fear particularly if we’re going to start restarting services. You want to make sure everything comes up, and so automating the patch almost like it’s step one, right? And then there’s building out the test afterwards to complete the process.

Steve: And that’s the interesting thing being on the DBA side of the world. Oftentimes you don’t always understand that entire application or web service or whatever it may be that’s using the database and know how to even tell if it’s working correctly. And I think that if you can have the tools that automated it that would really be great.

Robert: Absolutely and it definitely wasn’t something that we had done on our own. We had some amazing test people and developers for the application I worked on at Microsoft and they were really proactive about helping us build things like this and creating the web service call that we’ve made then do a dummy transaction in the database and make sure everything was working.

Carlos: It takes a village to do automated patching.

Robert: Absolutely. It often gets lost but we have to remember Dev Team, DBA Team, whoever we are we’re all on the same team ultimately and the success of the application really benefits all of us.

Carlos: Sure, it’s almost taking me back to the conversation that we have with Richard, Steve, about building trust our teams. This is kind of an interesting scenario where it gets to be a win win. One because we get to apply the patches that we want for security purposes but then application people can be assured that the application is still available when we’re done with it.

Steve: Right. And if you have that all automated then it’s less extra hours in the middle of the night for all of those people involved as well.

Carlos: Exactly, which I think everyone would agree is a good thing. I wanted to ask about deploying as well and restarting, you mentioned restarting the services. Is that across the board? I think with SQL Server we tend to it a little bit more just because SQL Server is running, we’re applying that so there’s file dependencies that sometimes get lock and things. Kind of by rule are you restarting your services after you patch them?

Robert: Typically, it’s easy to determine whether or not the automation process you can configure it to reboot it if reboot is recommended. You know, like some patches will say you must reboot to complete installing this patch. Other patches like say applying a CU to SQL Server won’t require you to reboot but then applying the new version of SSMS probably will require you to reboot. Most of the automation or the system center patching that we use gets that message back as to whether or not reboot as needed and then we’ll reboot as needed. You can also override that if you have systems that you don’t want to be rebooted automatically. For example, if you have a system that has a known issue where like maybe when it reboots all the drives don’t always come back online. I have seen cases like that where until they got configure out why some of the drives weren’t coming online they had configure it to not reboot automatically. But in the past without having that when we schedule for patching we to take into account that reboot maybe required and so we schedule Maintenance Window according to how long we think it will take if a reboot is going to be required. And so if you got the time, in the past we just install it automatically whether it says to be rebooted or not, reboot it. And there have been few updates from SQL Server, few CUs, or service packs where it gave me no message saying it needed to be rebooted, so we didn’t reboot. And then the next time I go in there to try and install an update, I then later go in to try and install the SSMS update and it says, can’t install there is a reboot pending. Like, why didn’t you tell me? So if you got the time scheduled and you don’t and you just want to play it safe. I’d say, yeah go ahead and schedule the reboot if you don’t foresee it causing you any problems.

Steve: Alright, so one of the things you mentioned there was the patching of SSMS or SQL Server Management Studio and one of the things that’s interesting around that is that, I mean it’s used to be the Management Studio was part of the SQL server install and if you updated SQL Server you would get your Management Studio patches at the same time. But now Management Studio is detached, running separately and with a completely different patch cycle and we’re getting much more frequent updates there but the other thing is it’s not a very smooth update process like a lot of the Microsoft products desktop applications for instance. You will get a message that says, there is an update required or come part of the Windows update process. But with Management Studio pretty much you got to always download it and then install that update. I’ve talked to a lot of people who, I don’t know, they kind of a little bit bothered at how annoying Management Studio is with monthly updates. And I’m just curious what your take is on updating Management Studio? Because some people I’ve talked to they come across to say, “Well, I’m not going to update Management Studio every month because I don’t need those new features.” But do people really need to update management studio often for security purposes?

Robert: With SSMS there’s really not a big security issue with the SSMS. Nobody is going to take away SSMS on your machine and be able to do bad things with it. Like they could took over your SQL Server instance. So there isn’t a security reason to be diligent about patching SSMS. But they are trying to work out a lot with the process and a lot of the recent versions of SSMS have had bugs in them. And there maybe things you don’t notice right away for specific features that you rarely use but you can definitely look at the update to see what they’ve done and determine if it’s something you need to do. But for myself every time an SSMS update comes up I update the SSMS I have on my desktop. I rarely ever use SSMS on the servers themselves. I like to have SSMS on the server in case there’s a critical issue and the only way that you can connect is via the dedicated admin connection. So I would like to have the SSMS on the server but I almost never use it so I don’t update the SSMS every time on the servers, every time an update for it comes up. But what I will do is when a new CU or SP comes out then I’m going to apply it to my instances. I’ll get the latest version of SSMS and I’ll install that as the same time as I do the CU and updates. I know that SSMS is on the servers themselves is at least as current as the last patch that was applied for the SQL Server itself.

Steve: Yup. Ok, great.

Carlos: Ok, that’s an interesting approach.

Steve: So I knew you mentioned accumulative updates there as well. It used to be like maybe 8 or 10 years ago that cumulative updates weren’t really as safe as they are today it seemed like, and that people would generally just stick with service packs when they are doing patching unless there was a specific reason that they needed cumulative update. But it seems like today cumulative updates are strongly recommended compared to where they were in the past. I’m just curious on what your feelings are on that or your experiences.

Robert: There’s used to be a big topic of debate amongst DBA’s whether or not to install CUs. A lot of people said, they always install the latest CUs. I was one of those that always said, “I only install a CU if it has something that specifically fixes an issue I’m having.” For the very reason that they didn’t get CUs used to not get full regression testing. When they were working on 2016 one of the things they change is that CUs not get the same full regression testing that service packs get.

Carlos: There you go, that’s the change right?

Robert: Yeah, so for me that really took that element of risk out of it. So now I will install a CU at my first opportunity to perform CU updates on the instances.

Carlos: I think that’s the effect that Azures had on the environment and that they need to be able to test those to support the Azure environments.

Robert: Absolutely. And everybody who owns Azure gets those updates automatically already so it was really one of the reasons say had the building the same level of trust for their CUs as they do for service packs because if you’re in Azure whether they’ve taken that choice out of your hand you’re getting the updates whether you want them or not. It’s really one of the selling points of Azures is that you don’t have to worry about the updates. It’s all taken care of for you and it happens automatically without you having to do anything.

Carlos: Right, also true. So before we go on with the other scenarios because I had a lot of questions on, so are you bundling your Windows patches and your SQL Server patches or do you like to do those separately?

Robert: We’re not bundling them, in that sense they are separate.

Carlos: Bundles is the wrong word. I guess what I mean to say, ok it’s patching weekend I want to do all my Windows updates and my SQL Server updates at the same time.

Robert: Yeah, so that’s really the kaitens we’ve gotten into now, and all of our production databases are on availability groups. The way we do it is we always patch the inactive nodes one weekend and do all the patching of the inactive nodes and then we failover our availability groups so that the patch instances is now the currently active node. And then we will run with that for a couple of weeks and then we’ll do the other CU. Unless there are some issue that we feel we need to update sooner we’ll do that another weekend a couple of weeks later and then throwback again.

Carlos: Ok, and then back into this next architecture if you will. I mean, so many people, they are stand alone where there’s clusters, now availability groups. And it sounds like if you’re familiar with the clustering idea that same concept, I can apply my service pack to one, flip it over, apply it to the other, and then either to flip it back or whatever your policy is there. That hasn’t change much with the availability groups.

Robert: Exactly. I mean with the availability groups you still want to patch your inactive node just like you would a failover cluster, failover and then do the nodes that are left that used to be active and then now inactive.

Steve: But it also sounds like either way with that whether it’s an availability group or a failover cluster that you’re delaying the time between when you patch one half of it versus when you come back and do the rest of them, so that you can have time to know if something went wrong and the ability to flip back to the old one if something did go wrong.

Robert: Absolutely, and because the way we do it where we break it up with the nodes to different weekends. If we do run into an issue like if there is a breaking change in the new version which is a lot likely to happen with the service pack than with the CU, but if there is a breaking change and the secondary can’t continue syncing data because there are some upgrade in the database that it’s trying to do but it can’t because it’s the secondary. The type of behavior you would see if you had a major mismatch between the nodes, you then have the option because the unpatch side is now the inactive one. Typically, you don’t even need to wait till the middle of the night. If you say, “Ok I’ve got to get this patch right now.” You’re not going to cause a major outage by now going in patching the secondary to make it match the active database.

Carlos: Yes, then you’re back in sync and hopefully that fixes the issue and they can continue to replicate.

Robert: Absolutely.

Carlos: Yeah, very cool. And then testing, right? We talk about automated deployments, right? Any specific test that you do around admittedly and I guess you know, maybe I shouldn’t admit this, but I think a lot of times my testing has been, “Oh, I install in a lower environments.” The proverbial, wait for something to blow up, and then if I don’t hear anything I move forward, right? Thoughts around patching testing?

Robert: That’s definitely one of the things I’ve always done is I start with dev and then move on to test and then move on to production. So like when the new CU or SP comes out I’ll install it on our dev machines the same day that we get it. And that gives us some time to see if any problems arise there before we go there. Also, usually if there’s a major problem with the CU or SP usually you hear about it within two or three days of the patch going public. Because a lot of people are start installing it on the devs environments right away and if there are problems they start finding it almost immediately. By the time we get around to patching the first set of production servers not only I’ve had it running on dev and test for awhile. A lot of other people have two and a lot of these people are very vocal when they find problems and really put it out on social media, which is one of the things I love about things like Twitter because you really, and some may post out, “Oh, this new CU is breaking my machines don’t install it.” Even if it’s somebody you don’t follow people will start retweeting it and you’ll see it pretty quickly on social media.

Steve: Yeah, that’s interesting because that really shows how valuable social media can be in that case where, I mean a lot of people look at social media in the workplace as a time snake in complete ways but when you’re using it for things like that it’s incredibly valuable. Ok, so one of the things I really want to ask about and it sounds you’ve done a lot of patching, do you have any horror stories either that you’ve seen happen or you’ve heard of happening where somebody tried to do a service pack or an update or a new version of SQL Server and something just went horribly wrong.

Robert: I have seen a few of those. There were some well known bugs that got hit in CUs and service packs where the upgrade would fail, the first part of the upgrade would fail but then not always but sometimes if there are system updates that have to be upgraded. You know, it will run a script to upgrade those objects and it puts the database in a state where if at that point it fails it puts it in a state where it won’t come unwind because the database is marked as in an upgrade. So if the CU or SP fails at that point it will uninstall the upgrade but it doesn’t revert that database and take it out of that upgrading state. It requires manual intervention. Obviously, one of the easiest ways to fix it is to, and we only know this because we’ve had to go through these issues where the servers won’t start.

Carlos: Life experience here.

Robert: Yeah, find the script, try to run, start the server with only master running and run the upgrade script and let it complete and maybe it will come late when you run it manually, if not it may give you a clear error, pin you exactly where it was. Last time I ran into this is because they made some changes to the job systems, the way they track jobs in MSDB. And part of their script was it deleted all of system categories by default and then inserted a bunch of new system categories, and had the IDs for those categories hard coded expecting that the likelihood that somebody has created a custom ID or custom category and that it’s using one of these IDs were going to use. They basically just assume that those category IDs wouldn’t be taken up, so we had custom categories that we had created and those IDs were in use and so the script failed because there is a unique constrain on the column. These were really kind of basic things we would yell at our developers if they didn’t check for if they’re going to insert hard coded values and not check to see if they exist already. But those get reported, they fixed those issues, and we haven’t seen that kind of thing…

Carlos: Click, yeah, record again. I apologize, Robert. If you’ll just go ahead and just pick up that story again talking about the change they made to the schedules or the categories then we can keep going.

Robert: Yes, so last time I encountered this issue myself was because they were upgrading the job system in MSDB which involved, their script would delete all the system categories from the categories table and then reinsert a bunch of new categories. The new categories was actually about as twice as many categories and they just made the assumption that the IDs, the hard coded IDs, they were inserting for the categories wouldn’t have been taken up by any user created category IDs. But we had created several categories of our own so when the script ran it had the unique constrain error because some of the IDs they were trying to insert were already in the table, and so it failed in the middle of the upgrade. So I found the script that they ran, ran it manually, got the error that it was generating. And I say, ok, so I went there. I dumped out the data from the table into something so I could fix it later, and then just delete, emptied out the table, and rerun the script and it completed successfully. These issues have all been reported to the team. I particularly gave them a hard time over that one because that’s the kind of stuff we would yell at our developers if they did. Like, why are you inserting hard coded values in not checking to see if they exist already?

Carlos: Or nothing else at least go to like 500 or something, right?

Robert: Right. But they took all that information in and they address those problems and these things never happened for a really long time. The only recent occurrences I’ve heard of these issues happen for people that are installing old updates. Like somebody is installing, “Oh look I just found out there’s Service Pack 3 for SQL 2005. Let me install that”, and opps it fails.

Steve: Interesting as you say that. I mean can we even get service packs for 2005 anymore? I remember recently somebody looking for one and we couldn’t find it anywhere on the Microsoft side.

Robert: Right, so you can’t get it through self service, you can’t find it yourself, but if you need to you can call my reserve support and they can get you old service packs.

Steve: Ok.

Robert: They are just going to ask you why you don’t upgrade. Obviously they are going to want to know why you don’t just upgrade to supported version. But they only keep the current versions of the updates online for you to download, so if you need really an old one you’re going to have to call support or I sometimes see people posting it for money. “Service Pack 3 for 2005, does somebody have a copy?” I see a lot of people are exchanging old updates that way.

Steve: Ok, interesting.

Carlos: Well, awesome, very good conversation, Robert. We do appreciate it and some good food for thought there. Shall we go ahead and do SQL Family?

Robert: Sounds good to me.

Steve: Ok, so Robert, how did you first get started with SQL Server?
Robert: You know, I’ve been in love with computers for as long as I can remember. But I didn’t go into IT straight away when I got out in the real world. I actually worked with the [term unclear – 36:58] disabled people for the first 10 years of my, that was my first career for 10 years. I worked with the [term unclear – 37:06] disabled people. But I was really longing for something different and still have that passion about computers. I’d started doing freelancing work, was doing Perl scripting and other types of web development. And eventually I decided, ok, it’s time to make the jump into the IT field for real. And so I started looking for position as a web developer. That’s the cool job that everybody wants though, right? Everybody come out of college wants to be a web developer. Competition was very stiff for web developers. I went to several interviews and ultimately I got offered a position as a database developer. And I thought, ok, I’ll do this for awhile and try to work make my way into web development world there. I worked for that company for 6 years and over the years, they were a very small company and their needs change and sometimes I would do web development. Sometimes I would do application developments, sometimes I would do database development. But we came to a point three years into my tenure there where our DBA had left the company 6 months earlier and they had never filled his role. And they were telling me that they thought they were at a point where they could make me a web developer full time. And at that point I realized I’d no longer want to be a web developer. I really enjoyed working with the databases. So I made a counter proposal, told them, at that time I was the head of the database development team. So I proposed that I take on the role of a DBA and still continue being the head of the database development team instead of doing web development full time. So that’s how I got my first DBA position. So that’s how I really got into it.

Carlos: Yeah, funny how things change, right, over time.

Steve: That point of enlightenment when you realized you don’t want to be the web developer. I can remember a similar point in my career as well. It’s an interesting time.

Robert: I like to tell people I’m a recovering web developer.

Steve: Oh, I like that.

Carlos: Now, of all the experience you’ve had in SQL Server and SQL Server is changing all the time. But if there’s one thing you could change about SQL Server what would it be?

Robert: What I would really like to do is there is some really horrendous features in there that I would love to see go away.
Carlos: Just take them out and shot.
Robert: Right. But that’s not likely to happen. If I had to narrow it down to just one thing, I would say get rid of all the features that have been deprecated for years.

Carlos: Pull the plug already.

Robert: I mean as much as I love database mirroring. I mean I wrote a book on database mirroring. Get rid of it. It has been deprecated for much longer than things are supposed to be deprecated. I mean there’s an extremely long list of things that have been deprecated since 2008, 2012. It is like get rid of these things already. We have people that are still using them and the logic they use is, they said that were going to get rid of it after two versions and they still haven’t, so it’s obvious that they are never going to. Why should I bother switching to the new stuff? People are still using the SQL 2000 system tables instead of the current catalog.

Steve: Right, right, so force deprecation. I like it. So what is the best piece of career advice that you’ve ever received?

Robert: I would say the best piece of career advice I ever received was that you have to be your own advocate. Early in my career this was something I really had a hard time with. I’ve never been one to go out and say, “Look what I did. Isn’t this great?” But one of the hard things of being a DBA is if you’re really good at your job nobody notices. People only notice the DBAs who are always fighting fires. If things are always breaking and you’re constantly in reactive mode trying to fix, people are like, “Oh, he is a great DBA because he is always fixing out problems.” And then there’s the guy who’s, his servers are not experiencing these problems because he is proactive and he prevents things from happening, so people don’t see all the work he does. And so the really good DBA, really proactive one, gets overlooked because everybody else in the company doesn’t see the fires that you’re preventing. And so they think you just sit back there and do nothing all day. Learning to speak up for myself and to publicize the things I do definitely was one of the best things I had to learn how to do.

Steve: Yeah, interesting. Yeah, I can see that how the firefighter reactionary DBA can say I fixed these ten crisis situations, where the proactive DBA it can be more challenging to say that I prevented these ten could be crisis situations, right? That’s really what it comes down to. That’s the difference between the reactionary and proactive. Ok, cool.

Carlos: Robert, our last question for you today. If you could have one superhero power what would it be and why do you want it?

Robert: If I could have one superhero power I want to be able to freeze time.

Carlos: Oh, freeze time, so almost like be able to walk through it where nobody else is moving type of thing.

Robert: Absolutely. Time is frozen for everybody else that I may be able to walk about and do things.

Carlos: Yeah, very handy when the boss is in your cube, right? When is the database is going to be up?

Robert: Exactly, so I think goes down and people bugging you, you could freeze time and get it fixed and then unfreeze and say, “It is fixed.” “What do you mean?” I think the challenge with that power though would be keeping it in the spectrum of superhero power as opposed to super villain power.

Steve: Yes, very good point. So many of those types of powers could be abused easily.

Carlos: That’s right. Well, awesome. Robert, thanks so much for coming on the program today we do appreciate it.

Robert: Thanks for having me guys.

Carlos: It’s been great.

Episode 111: How do you prepare for conferences?

Sometimes just getting approval to attend a conference is a challenge, but getting the green light isn’t the only thing you should plan for.  Sure, you want to learn some things–but a google search can teach you too.  In fact this week we are hearing about new features and functionality from the Ignite event–and I am catching all of it in my bunny slippers.

Why attend a conference then?

Conference season is upon us and Steve and I share some thoughts on why we attend conferences and some of the strategies we have used in the past.

Show Notes: SQLDP TS EP111

 

Episode Quotes

“There is an evolution from the time you go to your first one… after 10 you’re maybe a little bit more of a veteran there. But that doesn’t mean there are still not preparation steps.”

“We should all be looking to get something out of the conference and not just, ‘Well, I want to go there and learn something.’”

“Plan ahead, sign up for what you can ahead of time but also be ready to adapt to the last minute if things change.”

“We want to reach out to these people but we want to be careful about how we approach them and talk about something positive.”

“If you’re just trying to spam people that’s a different objective and you’ll be in a different circle.”

Listen to Learn

02:45   Today’s Topic: Conference Tips and Tricks
03:48   The right mindset in going to conferences
04:27   Find your people
05:11   Connect through social media postings
07:12   Check your plans and papers
10:34   Goals: Why are you attending the conference?
12:13   Finding your people, scenarios of networking with people
16:46   Never dine alone
20:03   Right place, right time, valuable personal interaction
20:35   Meeting Famous People: Do’s and Don’ts
30:02   What to do after the whole conference is over? Whose circle do I want to get into?

 

Transcription: How do you prepare for conferences?

Carlos: Companeros welcome to another episode, Episode 111. It’s good to be with you today.

Steve: Hey Carlos, this episode is on conference tips and tricks. How are you doing?

Carlos: Hi, I’m doing well. Thank you. Yes, conference season is upon us and everybody is talking about conferences and we figured, hey we have this podcast so why don’t we join the conversation?

Steve: Yup. Alright, but before we jump into that, do we have any companero shout outs?

Carlos: We do. Shout out to Douglas Kemp, he chimed in says, he enjoys the show and normally listens the day they come out. Of course we appreciate, let us know that there is somebody on the other side listening.

Steve: Yup, sounds good. So we also had one from Russell Johnson and he chimed in about Episode 109 on certifications. What he said was one of the things that he uses is the Brain Dump sites sort of the questions that people have remembered or written down or copied from the test to see what he gets wrong. Meaning, he had try them out and see what’s wrong and then he goes and learns it. Part of the point that he made on that was it’s not just about memorizing those one’s that you don’t know. It’s about taking it as an item that you don’t know and you’ve got to go and learn it.

Carlos: Exactly, and that’s always the balance, right? That’s always a tough balance because you just wanted to get to the certification, and sometimes you’re willing to push learning off to the side to make that happen. As you can get that review or get feedback, learning is always good thing and we all need to learn, continue to learn. Ok, so when this episode comes out, of course we are talking about conferences. I’m sure the Companero Conference will come out but you have today, today is the last registration day to register for Companero Conference so you’re kind of a bit close companeros. Let me just say, you haven’t registered yet.

Steve: Sign up now, or forever hold your peace.

Carlos: Forever hold your peace, that’s right.

Steve: Or until next time.

Carlos: Don’t say that, Steve.

Steve: No, we want to people to sign up now and next time I guess what I’m trying to say.

Carlos: There you go. Hey, maybe you’re 2 for 1 special. Ok, there you go. The shownotes for today’s episode are at sqldatapartners.com/conferencetips.

Steve: Or at sqldatapartners.com/111.

Carlos: So ultimately, again as you mentioned our conversation is around preparing for conferences and kind of getting the most out of conferences. I think there is an evolution from the time you go to your first one. And now you have, you know that first few then maybe you get 5 under your belt then you get 10 under your belt. I think after 10 you’re maybe a little bit more of a veteran there. But that doesn’t mean there are still not preparation steps. And as we were going through this and comparing notes before we started recording I think we recognize that there is some changes. Not every conference I’m going to prepare for the same way and some of the outcomes that I’m looking for in a conference are going to vary from year to year even if it’s the same conference.

Steve: Right, and I think the things we cover here, I mean they could apply to any conference whether it’s a weeklong conference that you have to go overseas for or whether it’s a SQL Saturday in your own backyard.

Carlos: That’s right. I think we should all be looking to get something out of the conference and not just, “Well, I want to go there and learn something.” I don’t think we are maximizing the value of the conference if that is all that we’re there for.

Steve: Right, and I would hope that everyone is there to actually get something out of it besides just getting out of the offices as well.

Carlos: Exactly! Oh man. Yeah, because if you have that mindset you’ll probably not listen to this podcast.

Steve: Oh yeah, exactly.

Carlos: Ok so for preparation. Preparation is before the event, when that preparation starts again can vary. But one of the things that I’m looking at, depending on what the conference is, but basically the people that you want to meet. Who is it that you’re looking to interact with? So these could be speakers, so you’re looking over session material and you’re like, “Oh, that’s interesting, I want to chat with that speaker.” But if they are posting social media information it’s basically they’re or connecting with those people who you think will be attending, and that you want to further or connect with or network with at the event.

Steve: Right, and I think part of that social media, the side to figuring out who’s going via the social media is also posting that you’re going and things that you’re doing and other people may reach out to you as well.

Carlos: Sure, for example Companero Conference we’re using the #compaconf just to let people know that you’re there. You may end up on other people’s radar as well. And social media is that great advantage because it allows you to connect with others or find out something about them before you actually get there and potentially even give you something to talk about. Like, “Oh, hey, yeah I saw your Twitter profile and you like to do…” whatever it might be.

Steve: I remember at PASS Summit last year, I posted that I was going and somebody who uses Database Health Monitor and he had been using it for quite a while had reached to me. Wanted to meet up and ask me some questions, so we met up and chatted then I ended up running into him probably four more times just as we crossed paths throughout the rest of the conference.

Carlos: Yeah, that’s an interesting perspective. Of course, we are going to flip it out and kind of look at his perspective, so he is a user of a piece of software. He reached out to the creator of that software and was able to have a conversation with him at a conference. That’s pretty valuable, right? Just a little social media like, “Hey, I’d love to connect.” Now, again, it takes two to tango and not every tweet to all the people are going to go responded to although it is amazing how available some of these people are. But again, kind of putting yourself out there a little bit and trying to connect you will be surprised who you can chat with.

Steve: Oh yeah, everyone is there just like you usually looking for someone to chat with.

Carlos: Exactly. So the next one, checking your plans and papers.

Steve: Oh yeah, this is one, you want to make sure ahead of time you’ve had everything is lined up and ready. And I have a sort of an embarrassing story to share on this. Hopefully it doesn’t prevent me from getting invited to any overseas SQL Satrudays in the future or out of the country SQL Saturdays. But a couple of years ago, it was around the time I was working on the Database Corruption Challenge, and I had plans to go to SQL Saturday in Vancouver, British Columbia, Canada which I live just across the border from there and it’s like usually about an hour drive with traffic so I thought, “No big deal. It’s all good.” I want to go and I was a speaker, and I was going up Friday night, and Randolph West who is a friend, he’s been on the show. He and I had a planned to meet up and chat because we knew each other through the Corruption Challenge. I man had the whole weekend booked out with SQL Saturday and dinners and whatnot. It turned out that when I got in my car and I got to the border, the customs official from Canada pointed out that my passport had expired.

Carlos: Wouldn’t let you leave, Steve?

Steve: Wouldn’t let me in, which is interesting because they turn you around at that point and you go back. This sort of this no man’s land between the two borders once you’ve left with the U.S. and/or Canada. I couldn’t get into Canada but then I went back to the U.S. and they gave me even more grief about my passport being expired. But they eventually let me in, and then I had to contact everyone at the event to say, “Guys, I screwed up, my passport expired”, and I didn’t realized it. I didn’t get to go to the event. I didn’t get to meet Randolph that time. I ended up talking with him in a later time, and then Mike fall. He eventually jumped in and did the last minute session to cover the session that I couldn’t do because I was not allowed in the country at that point.

Carlos: Oh man. Yeah, that’s maybe probably an extreme example but I think even like your hotel, summit obviously is the behemoth conference that it is. There are lots of other after parties or events. They have the game event this year, so just making sure that you have those reservations ahead of time because if you wait to the last minute like, “Oh yeah, I’ll kind of wait and see what happens.” The tickets will be gone and then you’ll be left like, “Well, now what do I do?”

Steve: But the flip side of that is if you do wait for the last minute look for things that are coming up and we’ll cover that a little bit later, but plan ahead, sign up for what you can ahead of time but also be ready to adapt to the last minute if things change.

Carlos: And this is kind of a cultural issue almost in our conferences. And I think summit particularly where the number of people that just go. I’m not trying to bad mouth, it’s not negative per se, but a lot of people just go just to hang out. They are not necessarily trying to take away sometimes the same things that we’re going to be talking about here because our next item is goals, right? What is the goal of the conference? Why are you attending? For some people it’s just, I just want to get like a family reunion of sorts. Just get together talk with other and whatever happens happens. For others, that may not be the case. We want to get out with a specific goal. I can actually remember the first summit I went to which I think was 2013 it was in Charlotte, so I’m on the East Coast. Like, hey you have to get down to this East Coast but other than it being summit and everybody kept talking how great it was, I had no purpose in going. That kind of reflected in what I then chose to do at the event because I wasn’t prepared, I hadn’t thought about it. I was king of constantly winging it if you will. You know, again, there are pluses and minuses to that tact and it’s really just kind of more of a cultural fit there.

Steve: Yup. You know that was my first summit as well, and I didn’t have a lot of idea of why I wanted to be there. I know, I’d publish the book on common table expressions just before that and the publisher had a booth there and I did a book signing. But that was really kind of the only non-summit type, I mean the only sort of extracurricular type activity that I did. I didn’t know about all these other events or things to and whatnot.

Carlos: Right. Now, this kind of then comes into the idea of, ok, well so now we’ve prepared or we fail to prepare and we’re now at the event. Now there are two modes of thinking a bit about this so we’re going to take both of them, and one approach is to find your people. So conferences are just getting bigger, right, at least we’re finding, SQL Saturdays I guess that exception. They are kind of that middle conference. We have a unique community there but for the most part conferences are like 5,000+ people. The idea that you’re going to network with 5,000 people over the course of three days is just ridiculous, so you need to find your people. Everyone is not your people. Some years that might change or the subject to the conference might determine who your people are going to be. I know and some cases I’ve had my people, “Hey, I want to get better at public speaking. Let me go find those people.” So the year I launched the podcast it was I need to connect with Microsoft people to get them to be guess on the show so that was what my focus was, connecting with some of those. And I spent probably most of my time in the Microsoft area just waiting for all the PMs and whatnot to come in because I want to introduce myself to every single product managers that was going to be there in that summit.

Steve: I can remember a couple of years ago when I did the Database Corruption Challenge, part of it was all the winners got a t-shirt, database corruption challenge t-shirt. And because a lot of them were from overseas and they were coming the past Summit I thought, well I just meet up with them and just make sure that I drop that off to them while I see them at some point. So sort of finding my people on that one was going out and finding all those participants and it as almost like an Easter egg hunt to find all these people that I had talked with before and say, “I’ll drop you a t-shirt when I see you there.” There’s only ten of them but it was a lot of work to find all ten of those people during the conference.

Carlos: Yes, I bet it was. And another way to do that is to be so bold as to officially organize something. Now we’ve seen people organize and the two that come to mind are Steve Jones and Andy Warren. It seems like they are always planning something and they have some specific niche. Another example is Andy Mallen, who we’ve had on the program as well. I believe it was last year or it may have been a year before but he organized an LGBT event. So again, kind of finding your people and he went so far as to actually throw out and event and say we’re hanging out. So that’s another way to kind of find your people.

Steve: And I think that there are so many things happening and so many people there that no matter what your interest are you’re going to be able to find other people that have similar interest in whatever it is that you’re doing.

Carlos: Exactly. But the idea is you have to decide who those people are going to be and then its effort to go and find them. Some of the other things that we will talk about, it’s fine to just go and kind of row with the flow but then you can’t then knock the conference for not providing what it is that you weren’t looking for when you really didn’t know.

Steve: Oh yeah, yeah, this is an interesting one because I can think of conferences that I used to go to like more than 5 years ago it always seems, like before I didn’t really understood the thing that I’d go to conference and kind of the conference get done and multi-day conference and I’ll always end up in a restaurant by myself or picking up some food, take it back to the hotel room. It was always just sort of like kind of boring at that point in the afterhours. I think that the whole never dining alone, there are so many opportunities to meet up with other people who are probably going to be dining alone as well and just get to know them a little bit. And I think like back you mentioned Steve Jones, and Andy Warren, last year they had just a dinner event where you show up and you meet people and they get like a group of six people together and everyone is going off to dinner together, and it was a lot of fun. I got to meet some people that I probably would have not met if it hadn’t been for that but it really sort of changes what you get out of the conference because you just get exposed to more people.

Carlos: A great example for me was I was at Ignite down at Atlanta. I was eating, met some folks down and say, “Hey, can I sit here?” Trying to strike a conversation and it just so happen most of them were getting up to leave as I was sitting down. So it was me and this other fellow, and then Lora Rogers who you may know. She is a SharePoint person she is very active in that community sits down at the table. I didn’t know her. We strike up a conversation, she is there, she wants to connect with other people. She is like, “Hey, can I sit with you?” You know, we start talking about SharePoint and I knew that I wanted to have a conversation on SharePoint and so I said, “Hey, are you interested in talking with me on my podcast about SharePoint?” And she said, “You know what I don’t think I’m the best person for that.” But she introduced me to Todd Klindt, and that’s how, and he came on the program and we had a great discussion with him. And so it was very interesting that again just in that conversation the connections that can be made if you’re willing to go out and then say hello and talk a little bit about yourself.

Steve: Yup, absolutely. And I think that the worst feeling at a conference I think is to be at that table by yourself or to be sit down at a table and realized that nobody at the whole table is talking. I always like to kind of break the ice a little bit and just say, “Hi, I’m Steve. Who are you?” and just get things going. It doesn’t always work but it helps a lot. Yeah, that’s the predicament wherein we’re mostly kind of introverts so nobody wants to go out on that But the reality is that most of us are scared to death. We sit down on that table, “Hi, I’m Carlos.” You know, it’s like, what’s going to happen now? And you have to embrace that awkwardness and as you mentioned it doesn’t always work out. But that’s ok because when it does it’s really nice.

Steve: Yup.

Carlos: And I guess that kind of leads us into right place, right time. I feel like this is a little bit the opposite of finding your own people, and maybe not the opposite but instead of planning and trying to find people this is more of the happy accident.

Steve: Oh yeah, and I think you’re going to run into people you know or people that you want to know that you would maybe never run into anywhere else. Like PASS Summit I think was last year I ran into a friend from Australia that I’d known online and through when I came back to the Corruption Challenge originally but I’d known, but we’d never actually really met in person before that. We got to sit down and talk business for like an hour and a half at the PASS Summit. We didn’t talk about technology at all. It was all about the consulting business, and it was the right place, the right time and hadn’t been plan we just bump into each other and sat down and chatted and it was awesome.

Carlos: Yeah, that’s right. So sometimes you have to know when to embrace that opportunity. It could come with a sponsor or with a speaker with another attendee, and so yet always tough to gage but if both of you are getting value on that conversation and it continues to go, you know, let that moment marinate for a little bit because getting to it would be very very difficult.

Steve: I think that so often at the conference it’s all around the schedule and getting that next session. I think that used to be the way I was but I’ve found that it’s always better to, unless it’s a session that you’re speaking at obviously or one that you really want to go to. It’s always better to maybe go in late or skip a session if there’s some personal interaction there that’s valuable because, I mean, you can always watch a recorded session earlier.

Carlos: That’s right, so the learning part is still available, right, that connection part may not be available again. Now when I met people, one of the things that I like to do, and I don’t do it for everybody. If you don’t see me viewing your profile don’t feel bad. But one of the things I like to do after I meet somebody and particularly one that I think I want to follow up with is then either connecting with or at least viewing the profile of that person on LinkedIn. One of the things I like about LinkedIn is it keeps that history. I can tell who, like where I was been and who’s looked at me so it makes that connection a little bit easier. I don’t think we get that same ability like on Facebook or on Twitter unless we’re going to follow them which again is another option. But with the information that’s available on LinkedIn I can know a bit more about that person from a business perspective which is generally what I am more focused on.

Steve: Yup. Yeah, and I think that connection so that you have a way to follow up later if you need can be incredibly valuable.

Carlos: That’s right. I guess I wouldn’t hesitate to follow people. Like, oh yeah we had conversation with that person, let me follow, and then again we’ll get to the after the conference we can make changes there.

Steve: But I think part of the worst case on that is when you connect with someone and they’re just pure marketing. Yeah, and they sort of just launch into you with their ad campaign and buy, buy, buy. I mean, that’s the wrong way to do it I think.

Carlos: That’s right. That’s always rough. In a that’s going to happen, you know, the value, the up side is so much bigger than a couple of people trying to spam you because you can always just disconnect with those people.

Steve: That’s true.

Carlos: You know, don’t limit your chances, which is going to happen, and again you just have to understand because it’s a social medium you’re going to have interactions that and not all of them are going to be exactly what you want. Another one, so putting work aside again can be tough particularly with a long conference particularly like summit. You’re going to do precons, that can be whole week. So checking in a certain time maybe put them in the schedule can be something but if you’re trying to balance the two you’re just hurting yourself in my mind.

Steve: Oh yeah, couple of examples there, I remember last year at PASS Summit we had a podcast recording that was scheduled and partway through the day I left, came back to my hotel room and we did the recording but that was something that I probably could have pushed off for a couple of days until we were done. But those kinds of things happen. The other thing I remember is a friend of mine at the conference. I saw him once and I didn’t see him ‘till like the end of the week, and I said, “Where you’ve been?” And he’s like, “Oh, but we’re having troubles back at the office.” And I’ve been pretty much the whole conference I’ve been in my hotel room working. And I just thought, wow, ok. I mean, obviously you got to deal with the issues that come up back home but hopefully you can get enough in place so you don’t have to do that because at that point the whole conference is kind of a waste.

Carlos: Right, and it’s funny, so there’s system down issues. That’s probably you can’t neglect but if your company has paid for you to go to the conference put your oath of office email on, right? I mean push back a little bit on that. Unless you get the “Thou shall do this”. It’s ok to say, you know what, yes I understand you having that problem I’ll come back to it. You have to say it nicely but put the tales out there so that they know, “Hey, I’m at this conference I’m trying to learn. I have the following goals associated with this conference. If it’s urgent enough to take me away from the goals that I’m working on right now then escalate, I’ll be happy to help you. Otherwise, try to push it off, and then the last piece that we have or topics for during the event is this idea of meeting famous people.

Steve: Hoo, how exciting.

Carlos: Yes, so more than likely and summit is again using that conference we have the superstars of the SQL Server community, and you’re going to bumped into them. So what do you do?

Steve: Or what don’t you do?

Carlos: What don’t you do, yes, maybe we will just start with that and admittedly this goes back, I’m remembering because I think this was probably now that I think about it, so Kendra Little published a post and this was probably just before the 2013 conference because I was reading about going to the conference and what to expect. You know, Kendra is one of those people that people would recognize and know the name. And she wrote an article about people coming up to her conferences and of course she has been putting out all this content but people will come up to her and just kind of say the weirdest things like, “You were shorter than I thought you were” or “Your hair is longer than I thought it would be” or whatever, right? And it’s like, huh. Now obviously we recognize or we feel like we know those people because they put out content. We read their blogs. We watch their videos. We feel like maybe we’re friends, and that’s fine and that’s great. We want to reach out to these people but we want to be careful about how we approach them and talk about something positive. You wouldn’t just go up to a stranger and say, “Wow, you are a lot shorter than I thought you would be.” To role reverse for a minute, that’s what you are, you are stranger and they are trying to make a connection, and if you insult them not a way to make connections.

Steve: Yeah, that’s interesting. I know I mean there are times that I’ve been to conferences and I think, “Oh, wow, I’ve seen that person online or on Twitter or whatever and I’ve been kind of like a little bit afraid to approach them.”

Carlos: Right. And we are all are, right?

Steve: Yeah, and I think that don’t be afraid but don’t be like super interrupting either.

Carlos: Exactly, and don’t expect to get an hour. You walk up to somebody and don’t expect him to hang out with you for the next hour. I think this goes back a little bit to the right time, right place. As you may just have to find a person at a point where they’re willing to socialize, right? If they are trying to get to a session for example all you may be able to do is say, “Hey, I’m Carlos. I love your stuff. Thank you for… it’s great to meet you.” That’s all you’ll get and then maybe in some other opportunity you’ll to do it a bit longer. I think some of those more social times. Again if you happen to find someone with an empty seat for a meal those are great times because that person is kind of bound at that point, and you can take your conversation in different places potentially.

Steve: Yup. Oh yeah, so then after the event and you’ve just come home, maybe you’ve recovered with some extra sleep in a day or two. What do you going to do next?

Carlos: Yeah, so this is I think where the most valuable nuggets come in. And of course the learning is important and we want to have goals and things that we will implement in our environments particularly for technical conferences. But one of the things I like to think about is, and again think about it from an outsiders perspective is whose circle do I want to try to get in to. Right, so assuming that I’ve reached to people on social media before, I’ve tried to find my people at the conference, now I’ve had interactions with them. If you haven’t had an interaction with them then you can just cross those people off your list. The idea of then getting into their circle now after the conference is not going to happen. But if you’ve had those interactions with your people, “Ok, now who do I want to keep with”, and then the idea of, “Ok, how do I lunch this conversation?” What things did we talk about at the conference, like the reference or maybe what additional questions do I have that would help begin the conversation.

Steve: Right, and with that I think though you always want to make sure that you’re respectful and you do it tactfully.

Carlos: You know, that’s right. And so I guess I’ve phrased that in a way of questions. What can I ask them? It’s the opposite of, “Hey, we met. I’m a great guy and I can do all these things and here is my résumé. What do you think about this?” That won’t help either. But it’s engagement process, asking the questions, “Hey, do you remember me?”

Steve: Yeah. “No, no, who are you, Carlos?”

Carlos: Then again if you can gage, you get one email, then you go for the second email. Again, how do I engage in a community, their community? When could we next interact? Things like that. Maybe it’s commenting on a blog or on a LinkedIn article or whatever it might be just to continue the engagement.

Steve: Yup, but not in a stalkery or spammy kind of way whatsoever.

Carlos: You know, that’s right. I mean, if you’re talking about establishing a relationship and getting inside of circle. If you’re just trying to spam people that’s a different objective and you’ll be in a different circle, right? In the do not contact file.

Steve: Yeah, and I remember one from awhile ago where somebody made contact. They had some technical questions, I helped them out. And the next thing you know, every tweet pumping their product was mentioning me on Twitter just trying to sort of call me out to get access to the handful of followers that I have. I thought, wow that really turned ugly fast.

Carlos: Block, right, you know.

Steve: Yeah.

Carlos: Yeah, that’s unfortunate. I guess we should remember right, that we are using social medium, we’re going to have interactions that we don’t care for. But we shouldn’t try not to have them because the upside is just too valuable, too great.
Steve: Yup.

Carlos: Anyway, companeros those are our ideas for conference preparations. Some experiences we’ve had along the way. Of course we have conference season is upon us and for those who are coming to the Companero Conference we hope that you’ll apply a little bit of this. We’re going to be doing some different things as far as the social component just from an organization perspective. In year one we don’t have a great way of putting out all the social information before hand but we will be creating a channel in the SQL Server community workspace in Slack to help communicate and some other things like that. We hope that you’ll continue to reach out and start chatting with other attendees. If you’re at the summit or any of the other conferences of course come up and say hello to us. We would love to engage and connect with you.

Steve: Yeah, we’re not famous enough yet to be completely overwhelmed by people saying “Hi”. So just come up and say “Hi” and tell us who you are and how you’ve come to know us?

Carlos: That’s right. But if you say, “Hey, that bald spot is a little bit bigger than I thought it would be”, you can just. Not the best opening line. Let me just throw that out there.

Steve: I’ll send those people your way, Carlos.

Carlos: Thanks, Steve. Ok, as always, companeros we want to hear from you. You can let us know. On our podcast page, we had a couple of engagements there. We haven’t had anybody lately. You can actually leave a message. We’ll probably introduce a new segment next week. We might try to a little bit more. The shownotes for today’s episode will be at sqldatapartners.com/conferencetips.

Steve: Or at sqldatapartners.com/111.

Carlos: And as always, you can reach out to us on social media. You can connect with me, I’m on LinkedIn @carloslchacon.

Steve: And I’m on LinkedIn @stevestedman and we’ll see you on the SQL trail.

Episode 110: How do I build trust with my team?

 

As you work with other people, you will experience resistance when you suggest changes and want to implement something new.  Why is this?

One podcast listener shared an experience where a problem was occurring.  The suggested a solution.  The idea was rejected.  No solution presented itself and the idea was submitted various times.  When nothing else was working, the idea was finally implemented.

The suggested change successfully helped the situation.  Victory!

Later, another issue arose and a similar suggestion was made.  The expectation was the suggestion would be well received; however, this was not the case.  The idea was still met with hostile resistance.

Have you faced a situation like this before?  Why wasn’t there an trust built because of the first recommendation?

Our conversation for today’s episode centers around the idea of building trust on the teams you work with.  As most our listeners are in the data platform space, we thought it would be a good idea to reach outside our community to folks we might receive the most pushback from–developers.  We are happy to have Richard Campbell on the show with us today.  Richard is the co-host of the .NET rocks podcast and you might recognize him from one of his many Channel9 videos.

We chat with Richard about how we can build trust and some common ways we attempt to show authority can backfire on us and actually cause more problems.  We talk about some of the ways those we trust have gone about earning our trust.  I know you enjoy this episode.

 Episode Quotes

“Mandates are one thing, execution is another.”

“Being successful is about persuading people to want to do it in a way that is usable by everyone”

“Position or power actually undermines your ability to persuade.”

Listen to Learn

04:07 What makes up trust in a team?
05:07  DBA’s responsibility in keeping the data safe and aligning business goals
08:37 Setting team standards, collaboration, being flexible to the needs of the team
11:52 How to be effective and create influence by giving value to the people in the team?
13:35 Becoming a consultant that takes away the tension of the problem inside the team
15:20 Understanding the whole person, the whole effort, the whole team
17:01 Attitudes when encountering failures
19:35 Other ways on how do you save yourself from quitting?
20:06 Why lunch is the most powerful moment of the workday in terms of improving trust?
25:30 Understanding the other team’s workflow
27:00 Discussion on best practice, showing the work, and getting an effective team
31:45 Team building activities and time together in strengthening the team
38:28 No level of position or power enhances the ability to persuade people to do something
40:46 SQL Family questions

About Richard Campbell

Richard the co-host of popular podcasts .NET Rocks (www.dotnetrocks.com) and RunAs Radio (www.runasradio.com).  Richard  is the founder of the Humanitarian Toolbox (www.htbox.org), an organization designed to let developers around the world donate their skills to disaster relief organizations by building open source software.

Today Richard is a consultant and advisor to a number of successful technology firms as well as the co-owner and content planner of the DevIntersection (www.devintersection.com) group of conferences.

Episode 110 : How do I build trust with my team?

Carlos: Richard, welcome to the program.

Richard: Thanks guys. I’m happy to be on board. Thanks for inviting me.

Carlos: Yes, all the other podcast that you’re hosting and have been on we’re happy that you came to our side of the pond and if you will and willing to chat with us a little bit today.

Richard: Hey, it’s nice to just be a guest because I know from some experience that hosting is hard work. You got to remember and keep track of a lot of things. I just have to answer questions like this is luxury.

Steve: Nice.

Carlos: How many podcast episodes would you say you’ve done?

Richard: Like little over 2,000.

Carlos: Wow, I was thinking closer to 1,200.

Richard: There are 1,465 .NET Rocks! episodes. But I came on board at show 100 so there is 1,300 and something. There are 545 RunAs episodes so that gets us up into the 1,800 range, and there is about 150 Tablet Show episodes.

Carlos: Oh, there you go. That’s right, I forgotten.

Richard: Yeah, so we are coming right around 2000. Now the Tablet Show was this little interval in 2011 to 2014 when it didn’t seem like .NET rock that much, and we hedging our bet but we were still doing .NET Rocks! but we are also building another show, just in case because it seems like Microsoft was moving away from .NET for awhile there. And then it was clear they weren’t, we just rolled the Tablet Show back into .NET Rocks!.

Steve: Ok, interesting.

Carlos: Well then we’re happy to announce this is your 2000th episode.

Richard: Yeah, ok, I’ll let you have that number. There you go.

Carlos: Well, so Richard, I think many listeners are going to know from the .NET Rock show. We actually have and one of the reasons we want to have you on the show today was a listener chimed in and had a question about trust and we thought that we might role play a little or talk a little bit about how this idea of trust works in teams in our environment. The reader actually goes to say, “I’ve been listening to your podcast, collecting best practices and I’m the DBA in an organization. I’ve only been there in a couple of months.” But they had the SQL Server, it was having problems. I was recommending changes to the SQL Server based on best practices, right, things we’ve been talking about in our program and others. The developers were like, “No, no”, because they’ve kind of managed the database, were responsible for that environment. Well, finally they relented because they were having so many problems and they had an improvement in that server.

Richard: Wow.

Carlos: That’s right, so everybody goes happy. Well, then he said, “Ok, well hey let’s adopt this as best practice and let’s go put this on server #2.” And they say, “Whoah, whoah, whoah. Time out!”

Richard: Don’t get crazy now.

Carlos: That’s right, and so his question was if I was able to show value and demonstrate that I could be of assistance why didn’t that apply to a more general trust of me and my recommendations ultimately. And so this kind of becomes then the basis of our conversation today, right, building trust among teams and particularly our audience is our SQL Server folks, and how they can work with developers like yourself to build that trust and to kind of move the organization forward. And so I guess we’ll start there and what do you think makes up trust in a team.

Richard: Well, looking at your particular scenario my concern always is folks look at one win and call it random chance. It kind of takes three wins before folks really like, “Hey you keep being right.” Like maybe there is an actual trend here.
Carlos: Sure. There is that saying that I like, “A data point of one does not a trend to make.”

Richard: Yeah, right. That’s kind of a narrow curve you’re looking out there, so it’s usual to build more points. I think it takes a lot of patience to be kind enough to allow people to learn themselves. So let them collect empirical data around their own mistakes. The challenge is to sit down your hands and not say, “I told you so.” At the same time I think it is also important, I think it is a trap we followed to is tech people is to keep talking about the practices around the technology rather than practices around them. The responsibility of a DBA is to protect and curate data irrespective of the technology involved. In the end this is an asset of the business. The developers are building software to allow us to bring that data in and perhaps to do some analysis on it, but ultimately the hat you really wear is the one that says, “I am responsible for making sure this data is safe and thriving and useful.”

Steve: Yup, and I think with that data that you’re keeping safe and thriving and useful is oftentimes considered by many of the organizations like their crown jewels. I mean without that they would have significant business impact if it was damaged or gone.

Richard: I hope that folks see that although a lot of folks don’t. That’s where, you know, when you talk about breaking down walls between teams. It’s not talking about technology; it’s talking about business goals, and so being able to speak in those terms so that people get aligned. When we really talk about that alignment we’re all pulling in the same way. It’s that we recognize, “Hey, our data storage is worthless without your software being able to feed into them.” So I don’t want to be an obstacle to your ability to put that data in. That data is only valuable if it’s usable and it’s safe so that’s why we have security and why we care about organization of the data and whatever form that looks like. It’s about making sure that data is actually valuable. You can’t succeed as a DBA without allowing your developers to be productive, that’s a necessary part of the equation. But vice versa if they are writing software that doesn’t store data effectively their software is not that useful either.

Carlos: We’ve talked about this on the show before kind of going from the CIO perspective or the managers they want the technology team to understand that the business value, what problems are we trying to solve and I think to you point sometimes we get sidetrack a bit by the technology and fighting over. You know, what we are or not going to implement when really we should be aligning what we do with those business goals.

Richard: And often we have pain too, scars. We’ve had a data corruption, or we’ve had a failed recovery, or we’ve had a badly behave piece of software where we… on a table. And so you can be a little jumpy.

Steve: Yup and I think some of that too comes from history, and I think with the example of this listener I think it was someone who was new to an organization. They probably had someone there before, I don’t know anything about the background but there is obviously some reason the developers were a little bit shy of making changes. Maybe they had a bad experience before that and it might take a lot of successful wins before that trust will be regained.

Richard: Yeah, and one win does not a trend made. Like, congratulations, you are walking the path but you got to keep walking it. And at least now you get to refer back to some success that’s already happened in the organization, some talking points. Blanket concepts around overall practices, those take longer to grow, and I think especially development organizations, you need to look at that from the perspective of how do you folks normally set standards around your practices because often we have a lot of teams where there isn’t a standard set of development practices. Essentially you’re going have to work from team to team to see how they like to interact, how they want to work with data, how they want to work with the system as a whole and essentially tailor a data access solutions to the team.

Carlos: So that sounds like something that might have to be, I don’t know, maybe not mandated but ultimately you have to be a little bit flexible. I guess maybe the first component of trust is being flexible to the needs of the team and ultimately being, which maybe against your traditional idea of a DBA, but being flexible to their needs rather than being quite so rigid in certain instances.

Richard: I think you need to be aware of how those teams currently work. If you do have an organization with say a strong enterprise architect group that does want a press standards upon everyone then talking to an individual team about a standard doesn’t make sense. You do want to go talk to those architects about how they developed those standards, how those measurements have done and how they are implemented so that you work within the system rather than inventing something new. And if it is team based, if we’ve got an environment where “agile” or we’ve got a variety of teams working in different ways there’s many ways that happens and works that way then it’s more of a team based. So I think there should be an awareness of how everybody works makes a huge difference in your ability to convince one way the other. I really want to emphasize this idea and you said this like, “Mandates are one thing, execution is another.” I’ve seen plenty of cases where we have a set of documents of this is how we build software and this is how we interact with data except for that part where nobody actually writes that code that way, those are great documents. You’ve got to actually look at how the organization actually functions as well as what it says. And being successful is about persuading people to want to do it in a way that is usable by everyone. That they are productive, you are productive and you achieve that real goal of delivering the data, so I would much rather persuade than mandate.

Carlos: Right. I think and so we kind of almost get into a soft skills discussion because this is again not about technology but this is about how to make friends and influence people.

Richard: I don’t disagree and I’ve done this long enough now in one form or another. It would basically come into conclusion that there is no situation where the technology can’t do this. This is invariably can the team do it? Yeah, this is marriage counseling.

Carlos: As a database administrator we think that we have kind of add some value to the team. First thing we want to do in establishing trust is really just getting away of the land and asking the questions, what are the problems people having, how do they go about doing things and then determining whether or not you can influence or make a change in some of those processes.

Richard: Absolutely, and so this is an old Stephen Covey’s 7 Habits of Highly Effective People thing. If you want to change anything, you want to influence anyone; first you have to understand before you’re going to be understood. So spending that time understanding how the team actually works. What motivates them, how they build their code, how they interact with data first and foremost that’s the first step because once you have a grip on that you can finally speak a language about how data is handled inside of the team and reflect that back on how you see it arriving in the database. That understanding is a step in trust too. The first hurdle to get over is this idea that they matter, and that they know you see them as people that matter.

Carlos: And that they are more important than the problem. It’s just kind of an area where the internet kind of fails us in the sense of we can publish this information out there, best practices, this is the way you “should” be doing things. But then we try to force that on people and say, “Hey, this is crazy. This is the worst environment I’ve ever seen. How do you guys work like this?” You’re not making any friends there.

Richard: Well, I’ve seen plenty of consultants where their names actually come from con game and insult. It is super easy to tear down other people’s work. The reality is nobody goes to work with intent to ruin things like today is the day I’ll make a mess of everything. That’s not the goal. I start with the fact that people generally have good intentions and then make the best decisions they could with the information they had at that time. Invariably when we look back we have more information and so we’re always going to see it differently. But if you recognize that you made the best decision with what you knew at that time and the constraints that you’re working in, some of which are going to be visible and some of which are not, then you’re a little bit kinder. I think the number one thing I’ve said most of the time when working with teams as an external is, “Guys, what I see here is very normal.” I have the good fortune to work with lots of different teams, there are always problems. The question is how well do we recognize them and what do we want to do about them. It’s just super normal. And just take that energy out of you’re not a disaster, this isn’t the worst thing I’ve ever seen. Like suck the air out of that talk about. We are all trying to do the best we could with what we had, here is where we are. We know we can get better. We can’t do it all at once. It doesn’t have to be immediately. These are incremental improvements overtime while maintaining the requirements, whatever they has to be productive and continue to ship features.

Carlos: Sure, right, so kind of taking out the Facebook approach if you will. Facebook everybody always post the best things, all the highlights and then you compare yourself to that and you’re like, “Ahh, I’m not good enough.”

Richard: No, and you always, out of the 150 store procedures you wrote you’re going to show me your best one but the other 149 exist. I think that’s part of this being a trusted team is a group of people that know we have good days and bad days. We have our strongest work and our weakest work but as a whole we deliver. Understanding the whole person, the whole effort, the whole team, that’s the value proposition. It’s hard to get there. We want to only look at shiny but we have to look at the hole.

Steve: And I think really the key to what you said there was understanding the team and understanding what the team is expecting. An example of that is doing consulting. We’ve work for a lot of different teams and some team I’ve work with, their number one goals is stability in uptime and if you’re pushing out a new feature and something goes wrong they would rather you abort the deployment of that new feature that means it’s going to give stability in uptime versus whether the new feature goes out on time. Others I’ve worked with, it’s all about we got to get the new feature out on time, and if you don’t get the new feature out, well there’s a problem.

Richard: Well, and often you’ll see those two positions inside the same team. The person who gets rewarded for shipping new features is really not worried about uptime and the person who gets rewarded for uptime features other worst thing that could happen to your uptime.

Steve: Right, and I think that you may get an organization and I’ve seen this where you say, “We try to do this but something unexpected happened. We rolled back and aborted that project or that deployment and here is the plan of how we’re going to take it on next time.” And some people respect that and other people look at it as, “Wow, you just slowed me down. You got in my way.”

Richard: Yeah, and you also find places where any kind of failure at all is such a big deal you never want to talk about it again. We cover up the fact that roll back happened.

Steve: Yup.

Carlos: Right, then we get into situations so analysis paralysis, so people are so afraid of making that mistake that they just don’t do it.

Richard: Yeah, deploying more software, making those updates would be the worst thing that could possibly happen which is just not worth the risk.

Carlos: Exactly.

Richard: I think it’s a challenge to get to a place, you know, we talked about modern environments. What we’re really pressing on is not know failures but minimum pain from failures and rapid recovery. Fail but don’t fail catastrophically that you’re able to be able to test it in the field. You know you can get back quickly. And it’s not a big deal to do that recovery that you know when you fail. It’s not like science pop out of the server going, “Opps.” How do you actually know there is a problem? If you have to wait until a customer complains then you know it’s been going on for way too long, so how are we measuring success or failure? Do we actually know when we’ve done it right?

Carlos: So thought around, let’s say again to our audience say database administrators we do have a couple that maybe managing other database administrators or team leads something along those lines but most of them I’m going to say maybe don’t have the authority if you will to start again say, “Hey, this is the way we should start doing things.”

Richard: And let’s be clear you are never going to have that authority. Get over it. It isn’t going to happen, so don’t count on it. Don’t wish for it, find other ways. You need to be able to get to that end goal of being able to have safer, reliable data that’s valuable to the organization. You’re never going to get it with the stick of power so you’re going to have to do it in a different way.

Carlos: Yeah, so I guess let’s talk about some of those other ways. Obviously we talked about reaching out when you’re on a team that is dysfunctional. I guess thoughts around how do you save yourself from just throwing up your hands and saying, “Yeah, we were not moving forward. I’m out.”

Richard: Well, and lots of people you either quit you go somewhere else, or you check out. Just tell me what you want. I will do whatever you want then nothing is my fault. And if you want to be more positive than that, more constructive than that
you have to start establishing some relationships and some trust that you have inside. And I would argue, you know, we described a very simple thing, understand the development process. Well, turns out that’s really hard to do because understanding development process needs interrupting a lot of people that are working, and they will be annoyed with you. You can’t do it during the workday. You have to do it around that and at the simplest level. The one window you’ve got is lunch. You know, when you give it workday the only time you really can talk to people where you won’t be take them away from work is lunch. And on top that, humans are hardwired to trust those they break bread with. Lunch is the most powerful moment of the workday in terms of improving trust. We just don’t take it seriously enough because we’re hungry.

Carlos: And how frequent it is that we meet people consulting for the first time over a meal. I think it would apply, I feel like maybe I have applied this in other scenarios but it takes them getting used to. It can be awkward, right, the first couple of times.

Richard: Absolutely, so just go and ask, “Can we have lunch together?” One of my measures of sort of the health of an overall team is to say who’s having lunch with whom. If devs are only having lunch with devs, and DBAs are only having lunch with DBAs, and ops guys have only lunch with ops guys. That’s an isolation. You actually want to encourage folks eating with each other. A lunch and learn where we talk through broad concepts on software and not a particular technology. More of where we want this application to go, where we want ops and data and dev to be together in a room. You want an interesting experience you put together three lunch and learns for each one of those team, all teams go to it as long as you’re buying pizza people are going to show up, right? Each team presents their view of an application, so developers are talking about the feature they’re building the stuff that they’ve deployed so far, the DBAs are talking about how data goes in, how data goes out, how data it’s currently organized, how it’s backed up and restored, operations talking about network traffic, how security works, how it goes to the firewall, because generally speaking the other team just don’t know what those guys do all day. I don’t understand why your job is hard. And so just having a moment where each of the other teams gets to see the folks that do the other piece and go, “Oh, that’s tricky.” The average dev has never seen a network diagram. Their network diagram is client, cloud, or internet server, maybe database behind it, so they reduce the operation guys entire job to a couple of lines. And the DBA guys to that cylinder on the drawing. There is more to it than that, and so actually starting to understand it a more deeply that’s the beginnings of trust, of some insight into the system.

Carlos: Can you think about all the effort we put in to just learning technology, you could take a lunch or a couple bunch of lunch, you mentioned three, to go to the pains of actually learning what your environment looks like.

Richard: And learning what your data storage environment looks like and what a recovery looks like. I want you to get to a place, I’m trying to banished the concept of that’s easy to everybody because everything is easy once you know how to do it. I want you to understand and remember these things are hard. And the side effect of that would you say to someone, if somebody ask you about a tech, oh that’s easy. You hurt yourself in two ways. One is you’ve undermined your value. Clearly you can do it because it’s not hard, and then same time you’ve also harmed the person you ask because this like, well you might be stupid because this is easy. Saying easy is just a terrible thing to say. There is no upside to it. Now, you don’t want to go, “Oh my god, that’s so terribly hard you’re going to die.” But it’s like this takes effort but it’s worth it. It’s a much more reasonable thing to say. All of our jobs take effort.

Steve: You know, I can remember being asked recently about a specific test that I have done and they phrase it as, this is what we want you to do. That’s going to be easy isn’t it? And how much does it take you to get it done? The administrator at a very high level without even exploring any of the details on it, and as soon as they said it’s going to be easy that makes it very challenging to be able to come back with anything that’s going to be realistic to their expectations.

Richard: Yeah, it’s terribly undermining, and not actually constructive. Either you lie or you lie, like you don’t have any good answers at this point. And even the process of estimating effectively is going to force people to actually understand more which often they don’t want to do. So just this encouragement to try and understand more to see what’s easy and what’s hard. I mean, there is another interesting truth which when we start poking at each other’s workflows there are skills that each team has that can make those other workflows simpler. Devs tend to solve stuff with code so often what they really understand a workflow on a DBA side there is like there’s some things we could do in code that would simplify that process for you now that I understand it. So there is often low hanging fruit between the teams that until we have some understanding of how everybody does their work, you don’t see that. We do have different view points on the same project. Once we see each other’s viewpoints you pick some simple things.

Steve: I think the key there is seeing the other person’s viewpoint, and almost like you’re in their shoes to understand what they are trying to say. And so often that’s what’s lacking. People just say this is what I’m going to do because I’m a DBA, or this is what we do because we are developers.

Richard: Well, we hide behind phrases like best practices. I don’t have to think about your need because I’ve got best practice. And again, it’s the other form of easy. You’re undermining the other person’s work, so why are you surprised when they resist you?

Carlos: Alright, that is an interesting question.

Richard: Let’s tackle it this way. We will only get to call it a best practice when all of us agree it is.

Carlos: Right, so the team itself has to agree on what the best practice is not you bringing in some external source and deciding for everyone, “Hey, this is the best practice.”

Richard: And it’s not like you have to invent from scratch. There is no problem with bringing something externally and saying, “Hey, I read this from this credible source. They recommend these things. How would this fit with our plan? Do you see value in it?”

Carlos: Right.

Richard: I hate coming with such certainty, and it’s a disease we have as technology people. We’ve had careers. We’ve had our time in school where we liked the result. We want the answer. We hated showing the work. Remember that, how much you hated showing the work. Here’s the bad news, when you actually want to influence people you got to show the work. And we keep coming with solutions and people want to understand the path. That’s why you present the best practice rather than a set of ideas that get to a result. One is showing the solution, the other one is showing the work. And without the work you can’t influence.

Steve: Yup, interesting. I can think of this specific example where I ran into that, where I was debating with a development team on the use of GUID as primary keys and the clustered index. Of course these were random GUIDs not sequential GUIDs, and I went and I found some blogpost from Paul Randall and Kimberly Tripp talking about this. And I went and examine them, wrote up an example that showed exactly what they’re article said and showed it to development team and then they came and said, “Oh, that’s just some blogger talking about it.” I mean everyone out there got an opinion on it. And the mistake I made was I didn’t set who Paul and Kimberly were in that example. They were just two unknown bloggers to this developer.
Richard: Not the guy who wrote DBCC.

Steve: Yeah, exactly. And I completely blew it on that one because I didn’t set the stage for where the data was coming from that I was using in my proof there.

Richard: And you can leverage that blogpost to map it to the problem, we’re seeing these issues. Let’s face it, like there is lots of controversy around randomly generated GUIDs. It’s not free. It has certain advantages for certain circumstances and it has a certain cost. Sort of upfront by looking it both, so the things to say, is this a cost we’re willing to pay to get this benefit. I do simply don’t believe in the one right way. I’ve had too many years writing too much software to actually believe there is a right way. There is the way that works at that time with the tools and skills that we have. We can get better but if there was one way through these journeys we’d all be taking it. The fact that a bunch of us have been successful in a bunch of different ways is sort of proof, and that no tool will save you. You take the same set of tools in a different problem and fail.

Carlos: Yeah, we look at that in data all the time, maybe like to make fun of the reporting tools. This new reporting tool is not going to help you unless you’ve got good plumbing, good infrastructure underneath. How many times do we try to do the same thing like what we’re talking about, the best practice or some other solution to say, “Hey, this new thing is what’s going to save us.”

Richard: Yeah. No, I want the big red button and then when I hit it, everything goes well.

Carlos: Exactly.

Richard: It’s a lovely dream, it’s just not true. In the end the tools cannot save you, it comes down to people in practices and since people make the practices it’s only the people. That’s all there is, you have to deal with the people that you’re working with. And if you understand them and see what their strengths are, value those strengths and put that puzzle together, put those pieces together so that everybody’s strengths are emphasize and their weaknesses are compensated for. That’s when you get an effective team.

Steve: So one of the things that I’ve seen in organizations were like management will try and get cross functional teams, operations and developers to work better together is they’ll do some kind of an outing that involves trust exercises and team building and that stuff. What do you think? That does kind of thing help or is that?

Richard: Yes. I’ll tell you why. It’s almost not the activity, it is the time together. In fact, the more mundane the activity, the better in some respects. Packing boxes at a food bank is not intellectual work. And the good news is then you don’t have to focus on it, maybe you’ll start talking to each other.

Steve: Or even if you took the whole group for an afternoon of bowling. It might have the same effect.

Richard: Yeah, same effect for that very much the same reason. And again, introducing food into the equation because it does plug into our brains to connect more, time together without the pressure on the work. Now, I’ve gone an angle on this as well. A few years ago help setup a charity called Humanitarian Toolbox where we build open source software for disaster relief organizations. In my mind it was, I know developers want to volunteer their time in a constructive way and use their skills. What I realized was it was also a very powerful team building exercise because yes we’re still writing codes but we are not writing the code from the office and we’re not on those deadlines. We’re in this volunteer mode. And so the dynamics of the team shift around and you sort of see each other differently. All of those exercises external to our regular day to day work, so the pressure of delivery, the pressure of a 40-hour work week are off and we’re in that different mode, that invariably can strengthen a team.

Steve: Ok.

Carlos: And I think some might complain now that you’re asking to do extra things.

Richard: Without a doubt, and this is part of understanding people. Look if you’ve got a 3-year old and a 5-year old at home you probably too busy to take a weekend off. Those kids need care. I think it is part of a good organization to recognize where our team members. How do we support them? On the other hand they love to have a weekend when they didn’t were constantly managing kids. So the fact that you’re offering some child minding as part of that equation at the bowling alley? Brilliant! How do I actually get the whole team to show up and be engaged? Well, if I understand them, I know what motivates them. I know what challenges they have so we can work on those things.

Carlos: And I think if we take the approach of I am making an investment in the team, I am trying to form those relationships and not, “Oh my gosh, another meeting,” just something I have to slug through that’s not helpful either.

Richard: The team building exercise will continue until morale… Really, we’re going to mandate that? One of the organizations I worked with, there was several people that were very bent out of shape about leftover food. They would take it down to the soup kitchen they would make sure utilize, and that to me was cute. Like there’s a strong social conscience in at least a portion of this team and so when you put energy into that, let that be more visible, it was very easy to say, “Hey, most of us care about this stuff. What would you think about doing an afternoon at the soup kitchen packing sandwiches?” It was already a thing where we’re
aware was important to many of our team. So validation there, add some time away, a good practice, doing a little corporate citizenship were our company did something collectively, you know, that was important. And the side of that was increased trust in the team, a shared set of values outside of today’s work. And some time to finding out who’s really good with plastic wrapping and who sucks at it.

Carlos: And then all the inside jokes that come along with that.

Richard: Exactly. You know, everyone of those inside jokes is a reminder of trust. Every nickname kindly given is a reminder of trust. That’s what good teams are built from.

Steve: Right, ok good stuff.

Carlos: Yeah, now I feel like we need to go to the soup kitchen, Steve.

Richard: Well, you’ve got to find the thing that works for your team, whatever that may be. Part of this is exploring the values that exist within your team outside of the day to day work. I’ve done Habitat for Humanity as well. My father was electrical engineer. I know how to do wiring so I was wildly productive and enjoyed myself. And everybody was surprised, I’m kind of a geeky guy. And then you give me a pair of lineman strippers and some rowmax, and I’m putting light switches in, surprise.

Carlos: Yeah, that’s right. And I think I guess, and again we’ve kind of been coming it from the vantage point while you may not have the ability to put those things together in your interviews and your annual interviews or when you’re asking for feedback, right, these are the type of things you should probably pushing for and making those suggestions. Even going back to the lunch and learn, I’m a huge fan, I don’t know why more organizations don’t do it, and I guess they look at this like a cost but again for a couple of pizzas. I mean, it’s well worth the investment.

Richard: Sure. You can do this ROI. My goal here is to make a stronger team; team is more productive for the price for a bit of food, it’s ridiculous. There is an argument about wanting to take a break, time away. But taking a break to think about something else is still a break. So it’s just what does that look like. I know a group of developers that what they did on their break was cycling. They loved riding their bikes. And so the fact that we would help organize for them a place for them go riding over lunch maybe huge difference to them. Again, valuing the distinctive aspects of those teams so that they can get stronger and get that level of validation that we care about the whole person, not just the 8 hours.

Carlos: There you go.

Steve: Yup. Good point.

Carlos: Yeah, Richard thanks for those insights. I mean so great points. Again, no easy button, right? But there are ways to go about.

Richard: And I guarantee you it is very persuasive in your own mind that someday I’ll get this promotion I’ll be able to make this happen. There is no level of power that will make it happen. CEOs all the time throw up team building exercises and nobody shows up. Just because you’re the boss doesn’t mean it’s just going to happen. You will never do this from a position or power. And I would argue that a position or power actually undermines your ability to persuade.

Steve: Yup. I would completely agree on that one. I’ve seen people get into that position or power where they believed they had the ability to command and totally loss the ability to persuade anyone to do anything.

Richard: They were probably more persuasive before they got the promotion because they actually try to persuade. Leaning on your title it’s never going to work. Just actually care enough to understand and help people understand you, you’ll get results. Nothing is fast, nothing is simple, it just takes time and it’s worth it because the improvements last. Once you get into this righteous cycle of a stronger team adding new members to the team, it’s easier to hire. People recognize a healthy, happy team and want to be a part of it. When you already have a culture that people value more people adapt to that culture. This is a righteous cycle, it gets easier over time. The first steps are the hardest steps but further down the path, you wonder how you were never there before.

Carlos: Right, I think using that title again, if the approach is different like you mentioned that idea, so there’s that saying, “People don’t care how much you know until they know how much you care”, kind of going back to the title right. If the approach then changes from caring about people to kind of getting things done, “Hey, I’m going to implement this thing”, then again you’ve lost, you’re kind of coming with the best practice again and you’re back to square one.

Richard: Well, it’s amazing how much people build like that because nobody uses this like, “Wow, that’s really well built. I’m sure you’re enjoying it all alone.”

Carlos: That’s right. Well, awesome. Richard, thank you so much. Shall we go ahead and do SQL Family now?
Richard: Absolutely.

Steve: Yeah, let’s do that. So what was your first experience using SQL Server?

Richard: Well, I’m the old guy, right, so I actually had the 421 installation on NT3 1, the non GUI version of NT. It was bomb proof, man. That was a solid piece of hardware. I think it was probably 486, maybe a 386 and just ran every day, all command line, total vintage; and the app I was writing was Visual Basic 2 with VBX library called SQL Sombrero.

Steve: Oh, I hadn’t heard that one.

Carlos: I like the name.

Richard: SQL Sombrero, that is like some vintage stuff. When we got to VB3 then we’d actually have RDO which was a data objects layer for talking to SQL Server, remote data objects but before that it was third party to actually get to SQL Server. But that’s old, that’s original client-server stuff, man. It was great great software to run. It was very straightforward but getting all the pieces put together and getting it working. That’s actually pre ODBC, that’s how old that is. It would be later that I would have ODBC errors.

Carlos: Oh man!

Steve: Yeah, there were no ODBC errors then.

Carlos: No ODBC.

Richard: There was no ODBC. Microsoft developed ODBC because they were trying to get traction to SQL space and everybody has their own drivers. The whole idea of ODBC was a universal driver. I had a customer once tell me, “I want you to get ODBC up all these machines because all it does is generate errors.”

Carlos: Wow, so all that time working with SQL Server and with .NET, right? SQL Server has changed quite a bit since those days, right?

Richard: Oh for sure, yeah.

Carlos: If there was one thing that you wish you could change about SQL Server, what would it be?

Richard: Well, even though I understand why .NET existed inside SQL Server I really wished I didn’t. I’m happy that it’s off by default. You should leave it that way. The number of scenarios where running .NET inside of SQL Server is a good idea are so vanishingly small. And of course Microsoft can never take it away because people have taken dependencies on it now. Like that’s going back to 2005, and their justification was this was a big data problem long before big data was really a thing. And today we would solve it in much more coherent ways but here we have this feature that just persists and it’s painful. It’s so dangerous. And anytime I have some a pop on the head, maybe I’ll just to turn on .NET. Don’t do it!

Carlos: Step away, step away.

Richard: Step away from the process, don’t touch it.

Steve: And I’ve never seen that work well by using .NET in the database.

Richard: It’s a very tricky thing to do and honestly I get so old now I don’t know it makes sense. There are so many better ways to solve problems that that was supposed to address than with .NET inside SQL Server.

Steve: Ok, so what is the best piece of career advice that you’ve ever received? You are not your customer. And I mean that from a perspective of your developers are not a DBA, and you’re a DBA, so just presume they are not thinking like you do. You’re not the customer, and so every time you think you know you’re wrong. You have to ask the question. You have to be influenced. You don’t use the software the same way the customer uses the software. You don’t operate the database the way that anybody else looks at the database. Your viewpoint is unique to your space. You might be able to talk to peers but your peers aren’t your customers either, so just stop fooling yourself. You have to ask questions.

Steve: Sounds like good advice.

Carlos: Richard, our last question for you today. If you can have one superhero power what would it be and why and why do you want it?

Richard: Well, I guess I want to fly. Doesn’t everybody want to fly? It’s almost a cut out
00:45:00 really. That way I could tell the airlines where to go and fly myself. I’m in metal tubes in the sky a lot and I’d like to do less of that and still get the chance to work with the people I get to work with, go to places I can go.

Carlos: There you go.

Steve: Isn’t that your superhero power Carlos, that you wanted?

Carlos: It is, yeah.

Richard: Fly or invisibility. I don’t want to do time travel because I don’t want to be responsible for anything, right? It’s really those three, that kind of great superpower.

Carlos: It depends on which plane I’m in, how much I can work with my laptop or my tablet right but that would be the one thing, the next thing to figure out once you can fly is how do I get to get some work done flying around.

Richard: Well, obviously that’s going to be enjoyable enough that we don’t care to do anything else. Never that simple.

Carlos: Well, awesome. Richard thank you so much for being with us, today.

Richard: Guys, tons of fun. Thanks so much for having me up.

Steve: Thanks Richard. It was a great time.

Richard: You bet.

Carlos: Yes, we do appreciate it.

Richard: My pleasure.

Episode 109: To Certify Or Not to Certify?

In episode 64 we interviewed Patrick Thomas, the program manager for the MCP program which includes certifications.  There continues to be quite a bit of discussion around certifications, so it only makes sense that we give our take on the matter.  Both Steve and I are certified, so this may give away some of our thoughts.

Certification is a tricky thing.  Much of the value you get out of a certification is dependent on where you on in your career and what your next step is going to be.  This is going to be different for every person, so this decision is one that only you can make.  This can make a decision on which certification to get and even when to get it a challenge.

Our conversation really revolves around what a certification will get you and what it won’t help you with.  Do you agree with our list? Leave your comments below.

Here is our list of Pros and Challenges with certifications

Pros
  1. Measure your skills
  2. Foot in the door
  3. Exposure
  4. Confidence

Challenges

  1. May not really measure your skills
  2. Not the only way to show skills
  3. Lots of prep work, with little way to gauge your progress
  4. Lateral Help
  5. Text Anxiety

Episode Quotes

“Having those certifications listed on your résumé may make you stand out as a little bit better.”

“It gives you exposure to more stuff that you may not have thought about trying in the past.”

“There are just a lot of features to know, and then try to create a test that can cover all of that is just daunting task.”

“The blogging… or the speaking at SQL Saturdays allowing you to create that portfolio, certifications are no longer the only way to do that.”

Listen to Learn

00:28   Introduction of the topic: to certify or not to certify
01:41   Companero shout outs
03:28   Updates about the upcoming Companero Conference
04:20   A little teaser of who will be next episode’s speaker
05:39   Show notes links
05:56   The pros of certification
06:10   Pro #1: Giving you an opportunity to measure your skills
08:30   Pro #2: Getting your foot on the door
10:51   Pro #3: Giving confidence to the technology workers and the employers
11:45   Pro #4: Exposure
12:28   The challenges of certification
12:37   Challenge #1: It may not help you measure your skills
13:48   Challenge #2: Your exam exposure and the employers expected exposure may differ
15:53   Challenge #3: Creating a test for too many features will be a very daunting task
18:53   Challenge #4: It’s not the only way to show your skills
22:18   challenge #5: There is a lot of prep work required that is very time consuming
26:16   Challenge #6: Lateral help
27:08   Challenge #7: Test anxiety

Transcription: IT Certifications

Carlos: Hello companeros! Welcome back to Episode 109. It’s good to be with you again.

Steve: Yup, it’s good to be here. This week’s topic is to certify or not to certify.

Carlos: Yes, as we get back in here to the studio, Steve and I are chatting today. We kind of came with this topic. We’ve talked a bit about certification in the past but we thought we would explore this idea a little bit and kind of talk about some pros and cons of why you may or may not want to certify.

Steve: Yeah, I think what’s interesting with this is there are a lot of pros and a lot of cons and a lot of varying opinions out there depending on who you talked to.

Carlos: Sure, sure and maybe we’re going to squash some of this conversation but the reality is that you got to have them. I mean I have passed certification exams, you’ve passed certification exams. You know, we might gripe about them a bit.  Our conversation is not going to, I maybe used the wrong phrase there earlier, maybe not why you wouldn’t certify but maybe some of the challenges you will face around certification, that’s probably a better model there. I don’t think there is too many people out there saying, “Oh, don’t do that.”

Steve: Right, but before we get into the details there. Do we have any companero shout outs to mention this week?

Carlos: Yes, we do have a couple of companero shout outs, one coming from New York, so Alexander [name unclear – 1:53]. Sorry Alexander. I apologize. I should have practice that before we went on. But you know who you are, so Alexander is a long time listener. Got us after a little bit or he reached out to us when we had that little pause and encouraged us to get back on the air and to keep going, and so we appreciate that.

Steve:        That pause by the way was our summer vacation. It wasn’t like us giving up on it or anything.

Carlos: Yes, that’s true.

Steve: We also had a shout out from Aaron Hayes, and he just wanted to reach out and say that he appreciates our time for the podcast and he always listens to it on his commute and it keeps him up to date with some of the latest stuff in SQL Server. Thanks Aaron we appreciate that and Alexander.

Carlos: Yeah, we actually appreciate all those comments. Steve and I were talking before we started here. You know, we push this out into the internet, we laugh and we joke and we have a good time with it and then all we kind of see is download numbers. And so it’s nice to get just a little comment even if it’s that. We hopefully could sometimes engage in longer conversations but just to have people let us know that they are listening we appreciate it.

Steve: It’s definitely good. We don’t always know from those download numbers whether people are enjoying it or whether they are just thinking we’re crazy or what.

Carlos: Yeah, that’s right. Hey, check out what Carlos and Steve said in this, you’re never going to believe it. Oh boy! And we should note, so Aaron is actually going to join us at the Companero Conference.

Steve: Oh, speaking of which, what’s happening with that now?

Carlos:    It is coming up, and companeros you have two weeks left to register. So next episode comes out, that would be the 13th, September 27th is out cut off day for registration to just allow us to get ready. Yeah, so we hope to see you there. If you haven’t checked out the website – companeroconference.com, we will be down in Norfolk. If you are a regular listener to the podcast I think we’ve gone through the idea, we have a great lineup of speakers, we’re going to be doing peer conference. We have structured and unstructured time. We’re going to be going out on a cruise and so we think we have put together something that’s compelling and hope you’ll join us.

Steve: Alright, and this week it’s just you and I on the episode, Carlos. But next week who do we have that will be joining us?

Carlos: So next week I’m looking forward to Richard Cambell, so the Richard Cambell from the .NET Rocks podcast. Probably you’re familiar with them. He’s been in lots of different areas. You see him a lot on the Channel 9 as well. They are involved with the SQL Intersection or Dev Intersection Conference, and so he’s kind of all over the place. We had a user or a listener requested idea for building trust on teams and so I thought, hey let’s reach out to a non DBA, a developer type, and kind of get their take. I had a conversation, it was fascinating.

Steve: Yup, and I think that’s an episode that we’ve already recorded but it will be coming a week from now. I had a good time with that. It was interesting because he is probably the person that was on the episode who has had more podcast and experience than anyone we’ve had on any episode.

Carlos: Oh, no question.

Steve:  Because he’s done I think something like over 2000 podcast episodes.

Carlos: That’s right. In fact, we claimed that that episode was his 2000th episode. We claim that number. Yeah, that’s a lot of podcast. So for this episode on certification, if you want to get to the show notes, the show notes for today’s episode is sqldatapartners.com/certification.

Steve: Or sqldatapartners.com/109 for the episode number.

Carlos: Yeah, so I guess let’s go ahead and jump into that conversation, right. So we have this idea of to certify or some challenges that you face in certification. So let’s go ahead and start with the pros. Number one, certification is giving you an opportunity to measure your skills.

Steve: Yeah, it’s interesting because looking at that it’s one of those that even just preparing for it you can go through and see these are the things, the topics that maybe included on the exam. And you can go through and figure out, “Ok, where I am strong and where I am weak. What areas do I need to improve on?” It’s sort of a good way to build out your experience or build out your knowledge so that you’re a little bit more rounded on that specific certification.

Carlos: Yeah, that’s right. I mean, certainly you’re getting a broad exposure to all of the features and technologies that are in the database. Most of the time they could get sold in a book, you could take a peek there. I think some of the prep tools are getting a bit better to give you a better chance to get your hands on some of those things before kind of putting them in your environment.

Steve: I think an example of that for me was several years ago when I took the 70-461 exam. And I went through the prep material ahead of time, the whole XML in the database was a piece that I hadn’t done anything with at that point in time. But after doing the prep work I realized, wow there is a lot you can do there. Now, I’ve done a little bit with it since then but it’s not one of those things that I would have necessarily just stumbled across in my day to day work.

Carlos: Exactly. I think that kind of goes back to this idea, again we’ve talked a lot about on this program is sort of marriage of the technology and the business requirements, and that is as you begin to become more familiar what your business is looking for, the problems that they have, right. How can the technology help solve some of those problems? I think is you are able to combine those two then you have a very successful, you have a very interesting package there. You can be very successful as you begin to connect those dots.

Steve: Yup, yes indeed.

Carlos: So Pro #2 is getting your foot in the door.

Steve: Right, so what exactly do you mean by that?

Carlos: Yes, so I think a lot of times, and I knew this is the case for me particularly when you’re new, having the certification allows you to be able to say, “Well, hey, I’ve passed an exam. I kind of know what I’m talking about.” And so as you look for employment it just seems like particularly in the DBA space, the data platform space certifications are important to employers.

Steve: Yup, and I think that seems that’s an area that’s important at that time that you’re writing your résumé or preparing to jump to that next position. Just having those certifications listed on your résumé may make you stand out as a little bit better than someone else who doesn’t have those on their résumé.

Carlos: That’s right. I think it’s particularly prevalent or most needful. Let me try it again, sorry, Julien. It is especially important when you’re talking about career changes, right? If this is in technology so for example you’ve been a developer and you want to come over to data platform or networking, same thing and system administration. Or if you haven’t been in technology and you want to come in. You know, certifications are that way. Again, if nothing else just demonstrate that you have the vocabulary to be able to speak the language.

Steve: Yup, and I think an example of that, I worked with a guy who have been doing mostly like tech support type work but he wanted to do some more database thing so anytime there was a database project he could take he did it but then he wanted to prove that he knew enough about database so sort of jump out of that support role so he went and did the 70-461 exam. Took him maybe a year of prep work to get ready for that from where he was at and he eventually passed it and sort of prove that he could play in that area.

Carlos: Right. I think that’s maybe another point that we should make is confidence there. We didn’t have that on our list, maybe we should make that as another point. Even I remember going back and talking it in Episode 64 when we were talking with Patrick Thomas who was over the Microsoft Certified Professional Program at Microsoft, and that idea of they are trying to give confidence to the technology workers, not only the workers but also the employers. And so I think again, we’re going to circle back to this and sometimes those goals or what they think of the exams don’t always line up but it is important that both of them are looking to them and again want to gain that confidence. And then our last one which we kind of touched on a bit already I think is just exposure. You know, we mentioned getting into the new features. What’s there I think particularly now as new things begin to come out. What’s in the database that you’re not using and could you take advantage of it. I think exams help to do that really well.

Steve: Right, and I think I kind of jump to go on that under the measure your skill section with my story there on XML. But yeah, I mean it give you exposure to more stuff that you may not have thought about trying in the past.

Carlos: So those are our pros and interestingly enough we have a few more cons, or may not cons again, but challenges. We should change it to challenges. So as much as we talked about measuring your skills, our first challenge is that it doesn’t really help you measure your skills.

Steve: Right, and I think that might be better phrased as it may not help you measure your skills. I mean, the fact is there is some confusion out there and there’s things people can do to get the certification without really knowing what they are doing. For instance, people can cheat by going out and finding some of the actual exam questions online. I know after I took one of the exams there was a question that I wanted to look into a little bit more and I did some googling on it. And I find out after I had passed the exam that wow there is this exact questions that are published out there. And really if you have a good memorization skill you could go through all of those and memorize them and not know anything about the topic and passed the test. I hope that doesn’t happen very often but I’m sure it happens occasionally.

Carlos: Right, kind of along those same lines is the idea that we talked about employers wanting the certification, so what the exam is going to give you exposure to and what your employer thinks a certification is going to give you exposure to maybe slightly different. They are not always in harmony.

Steve: Yup. Now, I can think of an example there where I knew someone who was newer to SQL Server they had been working with it for a couple of years and they went and passed the 70-461 exam, and after passing that the employer’s take on it was that they could now do everything, anything with SQL Server, and that person could do all the skills of someone with like 20 years of experience. In a conversation with that person we discover that there is a lot of things that you do every day that aren’t part of that 70-461 exam test. Like everything that’s covered in 462 for instance.

Carlos: Yeah, and this is kind where the blessings and curses. Even when we think about SQL Server, right? It’s great for us because there are so many features that come with that but then when we talk about SQL Server we’re lumping in SSIS, SSAS, administration, performance. I mean there are just too many tools in there, and then from an employer’s perspective they are like, “Oh, SQL Server certified. You should now understand all of these things.” You’re like, “Well…” Yeah, it could be a bit challenging.

Steve: Yeah, and I think part of the key to that is really understanding what that certification has covered and being able to convey that in a way that whoever is looking at it, like your employer, understand what that means.

Carlos: Right. Another one that I added here and that is with the new versions, there are just too many features. The database now, I mean it can slice, it can dice, it can chop your onions, serve your veggies and all the other stuff as well. And so it’s interesting I think, and again we’ll see how those certifications come out but the 461, 462, 463 are still going back to 2012, 2014. Even there were two versions removed from that they haven’t quite abandoned that and I think it’s going to be challenging just with the sheer number of features that we have. I don’t know if certifications are going to change. Maybe we have new exams, smaller exams, pocket exams. I’m not sure how that will come out but I think there are just a lot of features to know, and then try to create a test that can cover all of that is just daunting task.

Steve: You know, I think some of those features that they included in the test are almost like features that the marketing team for SQL Server wants to have covered so the people know how to use the latest and greatest rather than knowing how to use what they need to use to get their job done. And those can be very demanding on occasion.

Carlos: Or the ones like you mentioned even the XML that kind of continue to get in there every version, then you’re like, “Really guys?” Having said that I guess, I should say full disclosure. I have actually participated in with an organization that reviews the tests and we go through, so on the actual exams and then on the practice test, practice exams. And writing test questions is really hard. And so you have four people, that at least four people, on this call with the moderator going over a question and all the answers and we’re debating back and forth, and we’re looking at documentation and trying to make sure the wording is just right and then you change something and you’re like, “Oh well then that’s going to affect this other thing. And what if they think of this and how they are going to interpret that.” It’s very difficult and I think sometimes we might complain a little bit about how the exam questionnaires are worded. But believe they are spending a lot of time and energy to try to get those questions just right.
Steve: Yeah, they certainly are. I think that we oftentimes forget that when we’re taking a test or when we’re practicing for a test.

Carlos: Sure, right, because I think as we get to later, when we’re in there sometime all that time and energy doesn’t help us and that can be frustrating.

Steve:  Yup. So another one of the cons that we have on our list is it’s not the only way to show your skill and that there are a lot of other things that you may be able to do and maybe even be able to do with less of the time investment to be able to show your skills. There’s whether it’s blogging, or speaking, or taking on a side project, or looking on open source project or any of those kind of things might be just as valuable in showing that you know what you’re doing versus taking a test.

Carlos: Yeah, that’s right. I mean, our pros was that idea of getting the foot into the door. Right, having that on your résumé, that’s not the only thing that you could put on your résumé to differentiate you. I think we’re going to continue to see it more obviously in the developer space. We’re seeing it a bit more but even like Github, right, so what your Github handle and allowing employers to go out there and looked at the code you’ve checked in, you’ve participated in, whatnot. That is another way to increase your profile and I think that to our advantage and it hasn’t been too far in the past that I’ve actually need to work through recruiters and like try to find a job. Oh man, what a hassle that it. Like I can almost, the back of my throat is like having this vile taste in it remembering what that process was like. Luckily, that’s changing a little bit and so it’s more of extending your network and trying to connect with other people. That will be the best and easiest way, maybe it’s not the easiest way to get a job, but obviously the most successful path. And so all of these other things like Steve mentioned, the blogging and whatnot or the speaking at SQL Saturdays allowing you to create that portfolio, certifications are no longer the only way to do that.

Steve: Right, and I think that may vary greatly with every employer too. I mean I had a job interview I remember several years ago where it’s like, “We don’t want to see your résumé. What we want to see is your Github profile.” And that was it, that’s all they wanted to look as a starting point. And at that point in time I hadn’t really done anything publicly on Github but I’ve done lots of other public stuff but I just completely struck out on that one. They didn’t care about certifications but then you go into another place and they do care about certifications. Certifications are the only thing, you got to be a little bit more well rounded there I think.

Carlos: Yeah, exactly. It’s not a one size fits all. And the other component there is what type of organization do I want to work for, right? I think that plays a role into how you build your portfolio. What’s going to be attractive to them is also attractive to you. That increase is a likelihood that you’re going to go there and be happy, and be able to contribute to the team.

Steve: And you know, it’s interesting that interview that they only wanted to see your Github profile. I was a bit disappointed when that happened but then I later learned more about it and realized I’m kind of glad that I didn’t get that job after I learned more. Another of the cons on the list is really around there is a lot of prep work required that is very time consuming. Now, it may be people are at a different skill levels or different points in their career that that time consuming part may vary. But I think that no matter who you are there is some prep work required to go in and make sure you know everything to be ready for that test. And for some people that might be a year of prep work if they are new, for others it might be a few days or week of review to be able to just get up to speed on that.

Carlos: Whoah, a few days! That’s pretty aggressive, Steve.

Steve: Well, I guess it depends like if someone has taken the exams over different versions for the last 15 years they may have just to catch up in latest things before going and taking that next exam.

Carlos:  Well, it’s funny that you say that because that actually happened to me on the 461 exam. I thought it’s writing SQL, like how hard could that be? So my challenge there was I actually really didn’t study. That was the problem I used to, “Ha, I’ll go in there and I’ll just do it.” And you have to have a 700 to pass and let’s just say my score was in the 600 somewhere, right. I was like, “Argghh”.

Steve: Let me guess, did the merge statements throw you off in the exam?

Carlos: There was a group of questions that I guess shall remain nameless that I think threw me off, that ended up spelling disaster there for me. But then I’m like, “Oh, ok I got a bone up on that.” So I guess I made the backwards mistake of instead of using the test material to find that out I used the actual exam to find that out.

Steve: Yeah. Which you know, I mean that’s one way to do it because oftentimes you can get a retake that’s free or not very expensive.

Carlos: Or reduced.

Steve: And with that, I mean not all the prep material is all that is prep up to be. I think that I did some work for a book publisher a few years ago. That was on the prep work for the DBA series, the 70-462. One of the criteria that they had for everyone who was contributing on that book and video series was that you could never have taken the 70-642 exam because for people who would actually taken the exam they sort of lead in to like giving away exact clues of here’s how you pass the test. So in order to get around the legal hurdle of getting sued in some way they said, “Well you just have to study what Microsoft says is going to be on the exam and then we write a book around that.” I did eventually after doing the work on that book. I did eventually go take the test and it was much easier to take after preparing that book or work on that video work for them but I can see that there were things that are very different in the exam versus what you sort of expect from reading some of the prep material ahead of time.

Carlos: Right. Yeah, that’s one of the challenges with the prep material. I’ve used the books before and it’s like they just go over everything and you’re like, “Huh…”

Steve: And they have to go over everything because you don’t exactly what’s going to be on the exam.

Carlos: Exactly, exactly, that’s right. And then the way they get worded and all the other stuff I guess that’s another hurdle, but just a sheer amount of material that you have to go over is a challenge. Let’s see, so the next one we have is lateral help.

Steve: So really that’ sort of saying that in most situations it’s probably not going to help you in your current position but it may help you if you are jumping somewhere.

Carlos: Right. Or you’re looking for something, either promotion. It seems like it’s a plus. I mean, it’s not to say that education is not good. We’re specifically kind of honing in on the certifications. Obviously listening to this podcast is a way to increase your knowledge and what’s in your skill set. But yeah, it seems and this is just a generalization but it seems in our experience it’s been, I want my certifications for my next job not so much for the one I’m currently in.

Steve: Right, so I think leads to another one on the con list is test anxiety. And I think that there is a lot of great DBAs and database programmers out there who are just awesome at what they do but they don’t always do well on test.

Carlos: And the format of the tests are changing. I mean, they are difficult, right, I mean to put things in order. It’s like A through J. It is knowing some of the syntax. It’s very very tricky. Now, this actually takes me back to Episode 10, the very very beginning, Thomas Frank, who runs a blog for college students. We actually chatted with him about some ideas of preparing for exams and kind of dealing with some that test anxiety, and so that was kind of interesting. You’ll have to forgive that episode that was one of the very very early ones so the audio quality is not the greatest there. But if that’s something that you’re thinking about it may be worth taking a listen to that Episode 10.
Steve: So another one of the cons as well is you can just keep taking the test until you pass, meaning as long as you’re willing to pay for it. I remember another exam that I’ve taken different from technology here was I’m a volunteer firefighter and EMT, and when I took the EMT Exam, that was way more stressful and way more detailed than anything I have done with the SQL Exam. But the criteria with that was if you don’t pass it you’ve got to wait a time period before you can retake it, so I was extremely motivated to make sure I pass it the first time. Whereas with the SQL Test, if you don’t pass it go back the next day or the day after that and take it again. It depends on your budget how often you can take it I guess.

Carlos: That’s right and luckily so there was I think was the 463 Exam, the Data Warehouse Exam. Oh man, I took that one at least three times, maybe four times, so good and bad. I mean, some of that again is just you try and balance what you need to know with what you have to get to pass the exam.

Steve: Yeah, and I think that and maybe that’s an approach you could take. You could just go take it and find out what you don’t know and then fill in the cracks. But also if you do it that way maybe you don’t enough and you just keep retaking it and filling in the cracks you’re really teaching yourself what you need to pass the test, not teaching yourself what you need to be a rockstar in that area for instance.

Carlos: Exactly, also true.

Steve: Which might be the way to do it for 463.

Carlos: Unless Data Warehouse is your thing, right?

Steve: Exactly, yes yes.

Carlos: Ok, so I guess the recap. Our pros, we have measure your skills, foot in the door, exposure and confidence.

Steve: And our cons were around it does not necessarily measure your skills, not the only way to show your skills, there is a lot of prep work that could be time consuming, it may help you if you’re jumping laterally but not in your current position, those with testing anxiety may have the challenge with it, and you can just keep retaking it until you pass.

Carlos: So there you go companeros, what do you think our list? Of course we are interested in your feedback. You can drop as a line on the show notes page which is going to be sqldatapartners.com/certification.

Steve: And you know just to follow up on all these. I mean as we’ve been through the pros and cons, and we hit the cons as our second half of the list basically. I mean overall I think it’s a good thing. I just think that there is a lot of positives and negatives with anything.

Carlos: Sure, and I think we change cons to challenges, right? So we are not saying that it’s a con to take the exam. Again, we’re both in there, so just challenges. Yes, so what do you think of that list. We would be interesting in hearing from you and comments that you have. Of course if you have other ideas or things we should be talking about on this podcast please let us know. You can reach out to us. I am on Linkedin @carloslchacon.

Steve: And I’m on Linkedin as stevestedman. And we’ll see you on the SQL trail.

Episode 108: The Future of the Relational Database

After a brief hiatus, we are back on the air to continue the conversation and let me tell you–we have a great conversation lined up for this episode.  The discussion around what will happen to the relational database, and by extension us as administrators continues to get quite a bit of traction.  Even within SQL Server, we are starting to see more features that don’t fit the traditional relational mode and a podcast listener inquired about getting our thoughts.  As I thought about a guest for this episode, I didn’t want to get someone tied to a product.  They, like me, would be biased and I wanted to get someone a bit removed from the situation.

Our guest today is Andrew Snodgrass, the research vice president at Directions and we chat about the future of the relational database and what the future of the data environment we manage might look like.  I hope you will find his insights valuable as an outsider.  While we don’t get into the specifics of what databases are mostly like to be around, Andrew does give us administrator some ideas on what technologies we should start exploring.

What are your thoughts around the future of the relational database?  Join the conversation and let us know!

 Episode Quotes

“So these things have come out as a natural result of trying to solve a problem that we weren’t able to actually do with SQL Server.”

“I think what’s going to happen is SQL Server is going to be there for traditional structured applications and database.”

“The great thing is that we can operationalize an R script in SQL Server.”

“If you’re SQL Server and you want to look at big data… you’re going to learn Hadoop.”

Listen to Learn

01:41 Thoughts about database evolution and transactional system
02:05 Ideas about big data and database tools and services
05:40 Single platform for data: is it a fair approach?
12:00 JSON data, storage cost and compute cost
12:48 Effect on the Data Team
14:15 R script on SQL Server
14:46 Role of the Express version increasing in organizations
16:44 Azure services, data lake analytics, U-SQL on organizations
19:05 Technical skills that are robust in line with big data and data warehousing
21:33 Segway about Power BI
22:28 Azure vs Containers
27:13  Andrew’s advice for executives investing in their data platform
30:57 SSIS, SSAS, Power BI
35:56 SQL Family questions

About Andrew Snodgrass

Andrew SnodgrassAndrew Snodgrass is the vice president of research for Directions, a organization that help guides IT executives on technologies, strategies, product roadmaps, and licensing policies.  Andrew leads the research and analysis of emerging trends in enterprise applications (primarily ERP and collaboration tools) and database management technologies with a focus on Big Data, business intelligence solutions, cross-platform mobile application development, and cloud hosting services.  Her is currently and adjunct professor for the Albers school of business and Economics at Seattle University.

Transcription: The Future of the Relational Database

Carlos: Andrew, welcome to the program.

Andrew: Thanks!

Carlos: Yeah, we appreciate you coming and taking some time to be with us today. We have kind of an interesting topic. Ultimately, our audience, our SQL Server folks, and we think about SQL Server, you know, we think about transactional system. It’s been around for a while although it has definitely evolved a bit. But one of the common questions that we get is what about all these unstructured data and even in SQL Server now we are starting to see that influx sort of kind of with XML and now we’ve got JSON, 2017 is going to have graph databases, Python, R. All of a sudden we are starting to see this intersection of what we could traditionally think of unstructured and structured data commingling in the same environment that has a lot of people of questioning; and not to mention the rise of other databases, right? It’s 2017 and Microsoft comes out with a new database technology and they are not the only ones. There are databases almost sprout out as often as applications all trying to solve a particular problem. And so ultimately our conversation today is going to revolve around this evolution if you will and kind of where the transactional system is going to end up. So I guess maybe take us through some of your thoughts there and kind of what you’re seeing and maybe what some of your customers are asking you about the same idea?

Andrew: Sure, along that line I think some of the rise of the M structure, the big data, semi structured, all these stuff have come around naturally and I think that the introduction of database tools and other services and capabilities have come about because as traditional structured, traditional relational database guys keep trying to force fit unstructured stuff into structured databases. And so these things have come out as a natural result of trying to solve a problem that we weren’t able to actually do with SQL Server. And so I don’t know that it’s necessarily that we’re seeing any kind of decline in traditional databases or in the relational industry that we’ve always been dealing with. I think those types of workloads continue to exist. I think they are still going to continue to exist and you still got financial systems, you still got procurement systems, and these things lend themselves well to a traditional, relational structured database model. But they don’t do everything. One of the things that I’d always look at was how do we handle situations where you need some kind of lightweight database management system. And SQL Server despite some attempts to give us lightweight database management didn’t always do a good job of it and so when JSON came about, XML you could say a little bit, but really when JSON came about and the idea of saying let’s go mobile. Let’s push data out to self describing. These seem like natural solutions and now that we’re seeing SQL Server come back and say, “Well, we’re going to try to make so that you can tap those sources.” I don’t know that I would necessarily think that SQL Server is trying to be everything although Microsoft might like to portray it that way. I think that SQL Server is trying to tap everything. And so it’s one of these things where I’ve got my traditional product list, my customer list, all my transactions that are in a nice, structured environment. We are working an application that people know and trust and love but I need to be able to get into other areas. I need to grab data from other areas and push data out to other solutions. So they are putting these tools into our SQL Server environments so that we can interact with those things intelligently and more seamlessly so we don’t have to push to an ETL tool or we don’t have to do some kind of extraction and then modify the data to push it out somewhere else and then bring it back. We can kind of put all these things into one environment and control it there.

Carlos: We’ve definitely drunk the KoolAid. I’m Microsoft kind of through and through if
(00:05:00) you will. Is that a fair strategy as I go and try and talk with businesses? One of the things that I like about this idea is that at least it gives me a common tool set from someone who’s been in SQL Server now for 15 years. That I don’t have to go out and learn all these other platforms, right? Now, obviously there are some differences, R or Python, so there is a learning curve but at least from I know the data in SQL Server there are some administrative things that I don’t have to relearn. Things like that and I think it will be easier. But it gives me the ability on a single platform to have all of that data. I mean, is that a fair idea or a fair approach?

Andrew: I think a fair assessment on it is I think that from an additional enterprise standpoint, I think it’s very attractive and I like it personally as well, large organizations, medium organizations, anybody that has applications that they’ve developed over the years where they’ve invested. This isn’t just an investment in skills and application creation and the rest of that but they actually have servers and they have licenses and all kinds of other stuffs that go along with that and the idea of saying, “If I can take advantage of what I already have and expand on it instead of trying to recreate it. That sounds very attractive. Here is a great analogy on it. There was somewhat type of when we went from client-server to web-based apps. There was this need to recreate all these applications so they work in a browser and it would have been nice if we could have kept our applications to simply have more kind of browser. And that’s kind of what we’re getting at here is we have all of our data sources, all of our capabilities, but we want to extend it. And that doesn’t mean you need to learn how to write web apps and it doesn’t mean you need to learn JSON and convert data. It doesn’t mean that if you want to go to Hadoop you’re going to have to learn Polybase and understand how to do a scale out compute environment and being able to tap those resources and manage that data. You might need to learn some MapReduce at the same time. But this is learning additional things and not replacing what you already have. And that’s I think part of the attraction of beefing up SQL Server to do more than it was before.

Steve: And I guess the way I’ve seen this happen a few times that I have talked with different people is that they’re using SQL Server, they’ve been using SQL Server in their shop for years and then they realized. “Ok, now we want to do unstructured data.” Big data as they often times end up calling it. And then they decide, “Ok, well we’re going to go with Cassandra or DocumentDB or CouchDB or DynamoDB.” Or one of those that is totally separate from anything that is SQL Server. I find that sort of the story goes with those oftentimes is that it’s kind of a decision. Are we kind of keep doing SQL or are we going to go with this other database or are we going to do a merge there or sort of using both sides of it. IT sounds like what you’re describing here is taking advantage of what’s being added into SQL Server so that we don’t have to go and look at those other technologies. Is that where you’re going?

Andrew: No. Actually, Steve, I don’t think that’s kind of the case. I think the merge is what’s going to happen ultimately. It’s kind of like buying a stereo system that’s an 011. You basically get a very generic set of components. If you really want the best you buy separate components. And I think what’s going to happen is SQL Server is going to be there for traditional structured applications and database and stuff but people are still going to be looking at Cassandra, cloudera, Hadoop, any of these for a true multiprocessing, parallel processing capability that they really don’t want to spend the money to try to make SQL Server do. The idea of that is if you go to a bid data solution you have a distributed computing environment with a lot of servers out there. Each with some local storage so that the computing the query happens locally on the data and you use parallel processing to do a lot queries across a lot of unstructured data at once to return the said information. If you try to do SQL Server to do a
(00:10:00) scale out and that same type of configuration your licensing cost alone is going to kill you. It doesn’t make any sense to say that you’re going to have 50 or a thousand servers each running a SQL Server Standard/Enterprise edition license. Oh my god, I can’t imagine the amount of cost that you’re going to have there. So what you’re actually looking at is you got a SQL Server environment that’s very powerful that has some really good SQL Server capabilities on it that uses Polybase to go out and tap that unstructured data that’s actually in a proper Cassandra, Hadoop deployment. If you really need to get in big data you’re going to have to look at a proper big data solution to host the data and run the compute. But you don’t have to run those stand alone. You don’t have to just do a MapReduce to get a data set that then you have to go through Integration Services or some other ETL in order to get in what your structured data. You will be able to do that directly from SQL Server so that you can query your structured information, your product list, your customer list and you can do sentiment analysis against a bunch of unstructured data out in Cassandra and bring those together in real time. That’s where I think the merging and the intelligence is going to be.
Carlos: Interesting. I think kind of with that a common point is like IoT, right? So you’re collecting, so all these sensors have all these information. SQL Server probably may not make the most sense for the repository for all of that information, right? But you have these sensors, you know, caches that on the web which you can then pull in to a Hadoop cluster or whatever else and then start to do that aggregation or some of that analytics on it and then pull in kind of the summary information and that maybe what you store in SQL Server and then of course then there is the reporting on top of all of that.

Andrew: Right, and that follows, Carlos, in the same line with the idea of JSON. It’s great that you can some native JSON stuff within SQL and you can push data out and pull data in. But from a storage cost and compute cost do you really want to store JSON data in blob columns within the SQL database? You are going to run some cost and performance issues that you might not want to deal with so I don’t think that’s necessarily the road with the JSON side either. I think store it where it make sense, where it is optimized and where it belongs. But I want to do stuff with that data and I want to do it in real time and that’s where bringing it into SQL Server and attaching it to business data make sense.

Carlos: So then let’s talk a little bit about the effect, are the data team just going to get bigger? I mean so we have like cache and some of this other caching layer that the programmers are already kind of doing. So now we start adding Hadoop and we’ve actually done an episode on Polybase. Now Polybase is still kind of an emerging technology. They kind of came up with the v1 but those worlds are still kind of very different so does that mean that I’m now going to have a SQL Server person, I’m going to have a Hadoop person, I’m going to have a Cassandra person or are you seeing organizations say, “Ok, guys. Now, it’s time to go learn Hadoop and you’re going to be responsible for both.”

Andrew: In some, and it depends on the size of the company. The really intelligent ones are investing in data architecture. It doesn’t matter just of saying we want big data you know everybody says that. How do you do that and how do you do it without spending too much.

Carlos: So the plumbing, the connections have become very very important. How’s that data is going to move and flow.

Andrew: Right, and so and the same thing is true now with machine learning. It’s great to say that we have R and Python in 2017 but what are you running this against? And what type of performance that you’re going to take on your servers with it. The greatest thing there isn’t suddenly we are able to do R scripts within SQL Server. The great thing is that we can operationalize an R script in SQL Server. And it’s that I’m going to R Studio. I’m going to build my script and I’m going to test it out in my data. I want to make sure everything is the right way and then I’m going to put this into a stored procedure and make it part of my application so that it performs the R script on the data, the right data, at the right time and then gives my users something useful.

Carlos: Exactly. I think now, so having said that do you see the role of the Express version increasing in organizations? I feel like adding some of that functionality
(00:15:00) to Express was a direct result of the rise of open source and particularly MySQL. And so to that idea now that I have the capabilities, and when we start creating these models maybe we don’t need 10 years worth of data. I need a month’s worth, a quarter’s worth. And then I can push that off unto an Express version let my data scientist or analyst or to whoever run all the items they want and then once they’ve decided, “Oh, I think this is where we want to go.” Then again kind of that integration, that’s in SQL Server already I could just move that into my production environment and have a more seamless experience.

Andrew: That’s a really interesting question on that. Most of our customers tend to be larger enterprises and so they have their MSDN Visual Studio subscriptions and even the data scientist they are putting MSDN Enterprise edition on or Standard and running it from there. I don’t know any answer for that. I think you probably could.

Carlos: Sure, the limitations that you’re going to run into. You know, 10GB or whatever. But again it’s that I can move that off my environment with the flexibility I get the tools so I guess I’m kind of interested to see how that plays off or plays out. Obviously, if you are Fortune 5,000 Company maybe that doesn’t makes sense but for these smaller organizations they have a couple of people trying to work on this. It will be interesting to see what happens I think.

Andrew: You know, on that line opening the door on that one for some of the may have customers both large and small we are seeing the use of Azure services spinning up capability there, moving data out, testing it out and then scaling down is a way of getting around local deployments and testing. Like Azure SQL Data Warehouse, all of a sudden you’ve got somebody who has 10 or 15 years or 20 years with a historical equipment data or performance data and they push this up to an Azure service and then run one of the analytics services against it. You know, they are doing some machine learning and some bigger analytics that they would never attempt on premises.

Carlos: Sure, because once they’re done they could just toss that out or spin it down.

Andrew: You know, and along those lines you’ve got Azure Data Lake analytics which was supposed to be originally just for going against the Data Lake storing environment with a bunch of unstructured data and all of a sudden they said, “Well, you know, we could tap into SQL database. We could tap into blob storage. We could tap into data warehouse.” And Data Lake analytics is all based on U-SQL and so we’ve got a really nice environment here for people that are comfortable with traditional SQL Server query capabilities to go out and run some really cool analytics and learn just a few new functionalities that are required for U-SQL and you got a scalable system out there to run these queries on that you don’t have to do on premises.

Carlos: Yeah, it’s very cool. I still think that some of that is like you mentioned for the larger organizations. I think as we start talking with people, I mean, there are those who are putting quite a bit of time to learn the U-SQL and take a look at Data Lakes. And I think obviously the bar is lowering but that’s still an investment right to take the time to learn some of those things and understand how they work. I guess like everything, all the technology continues to evolve and change.

Steve: Yup. It’s interesting because when we had our podcast talking about building your technical skills few weeks ago, one of the things we talked about was sort of the type of things that you can learn that stick around for a long time and the type of things that you can learn that are maybe a little bit riskier but they may go away or they may change rapidly. And it seems like some of the unstructured and big data and data warehousing pieces today are kind of within that riskier category because they are still sort of changing quite a bit. I mean are there any of these that you see that are kind of more robust to the point that or more mature might be the better word to the point that they are not changing quickly
(00:20:00) these days?

Andrew: Yeah, actually. Especially with the folks that are going to be listening that are SQL Server folks. You know, along that line if you take a look at Microsoft’s database landscape and say you’ve got structured, unstructured and semi and everything in between, the two that you would go into that are not SQL Server would be Hadoop and JSON. Hadoop is underpinning so much to the big data investment that Microsoft does into that if you want in SQL Server you’re probably comfortable to start learning Hadoop and understanding that environment. That was the whole concept behind Polybase was to touch Hadoop. If you actually look at the Polybase scale out concepts, this is a Hadoop cluster methodology for doing compute nodes. If you look at the Azure services you got Data Lake which is a Hadoop environment on the backend. You got Cosmos DB which is actually a Hive environment in the back end, and then you’ve got HP Insight which is multiple Hadoop type clusters that you can choose from there to do any kind of scale out environment you want based on the type of workload. Yeah, if you’re SQL Server and you want to look at big data and you want to say comfortable with Microsoft you’re going to learn Hadoop.

Carlos: Ok, interesting. There you go companeros. You’ve been warned. You heard here first.

Andrew: Yeah, and while you’re at it, learn some Power BI because Microsoft is investing in that.

Carlos: Oh yes, that is true. What it’s interesting I think that still kind of goes back to the plumbing. We’re actually starting to see traditional SQL Server people/folks trying to spin up Power BI environments and you’re like, “Hey, really cool reporting tool.” You know, the functionality is there, web browser and all that but then tickets are a little too big and they are like, “Hey, how to make this go faster?” And you’re like, “pull that out .csv file.”

Andrew: And then you do a Power BI premium where you have dedicated capacity or you end up doing a Power BI report server on premises.

Carlos: Yeah, that’s just coming out and it will be interesting to see that integration with Reporting Services that continues to evolve. I wanted to circle back quickly if we could because you mentioned Azure. We’ve actually talked a little bit about containers and containers particularly in the Linux world has been all the rage but it almost seems like maybe, I don’t know if competition is the right word, because ultimately the technology that they can give are different. Well, Azure versus containers. So if you’re a large organization and have an investment or haven’t made the move to Azure to that whole spin up for the way idea. I feel like that’s what containers are also trying to provide you. So is it an on premise Azure competition there when we talk about containers?
Andrew: Like yeah, that’s an interesting one. Maybe I’m too old struggling to find out a really good set of examples of doing containerization of apps and saying we’re going to put this out here especially when it is database related app. It seems like there is an off light you’re going to put in that container in order to make it work. But you know it’s interesting in Azure right. Microsoft’s whole move towards Linux anyway with SQL Server 2017 coming on Linux and the containers and all the rest of that. Azure has the Azure container services which has been out there for a little while, a year or a year and half, that allows you to spend up VMs and containerize things and then everything in it. But they’ve also just released a thing and it’s in preview right now called Azure Container Instances. And this is one where it’s a true platform as a service type. They provide complete control in the infrastructure so you don’t know the backend VMs. All you do is you’re seeing computing resources. I mean this many CPUs, and this so much RAM and here is my container image. Pow! Put it up there. Scale it as much as you want, run it as long as you want and then kill it. You don’t worry about the VMs or any of the backend stuff. You can throw SQL Server into this, and you don’t know if it’s Windows or Linux and doesn’t really matter because there is web access to it anyway. You get the URL and you just run it and you go for it. And so that’s another one where Microsoft is just trying to say we don’t really care on the
(00:25:00) platform. We know it’s the service you want and so here is how to go about doing the container and you don’t care what’s happening on the backend side of things.

Carlos: Now, I feel like the best example, and so we had Andrew on the program, different Andrew containing about container. The thinking was and I guess I’m inclined to believe it is that, at least from a database perspective. I can’t speak to the apps. You could do lots of different things with the apps but the best or the classic case for containers in the database is a development workshop. If you’re building like .com type of thing, right? If you’re a actively doing development that whole idea of, “Ok, once you’ve published or once you’ve pushed code. Ok now, let me take that snapshot, create a container, push that down back to all the developers.” So now I know they are developing against what is in production and it helps eliminates some of this migration issue.

Andrew: Yeah, that sounds very internal.

Carlos: I don’t think we’re quite, although I have heard a few people talking about it, I don’t think we’re quite saying that we’re going to throw containers at production databases just yet but.

Andrew: Yeah, you know, sometimes I hear so much about containers right now and it takes me back when we first heard about big data and then about artificial intelligence and it’s like, “Oh, I’ve got to get me some of that. I don’t know what I’m going to do with it but I got to get me some of that.”
Carlos: That’s right, so having said that what your ultimately with the directions, your organization, your consulting CIOs on technology choices. Now, obviously your mileage will vary right and it depends. It’s going to come into here but what’s your, I know we covered some of this pieces, and I guess you’ve already mentioned if you feel comfortable with SQL Server, you feel pretty safe gambling on Hadoop, not gambling but learning Hadoop. You know, taking the time to that could be your next foray. What else are you advising these executives in making investments in their data platform?

Andrew: So the things that we’re pushing or talking people about now is really along the lines of how do you take advantage of not only big data and machine learning capabilities…
Andrew: There you go they are implementing IoT and so that’s one of the things that we tell the folks is not necessarily to jump in because these guys aren’t going to do major investments or major overhauls. These are like organizations that have to plan out years in advance of what they are going to be spending their money and then get authorization for it and have to justify it. They’re getting a lot of CFOs and other CxOs that are reading too many magazines on airplanes that are coming back and saying, “Why aren’t we in big data? Why aren’t we doing artificial intelligence? What about IoT for our organization? How come we are not doing better analytics and all these stuff?” And so we’re trying to show them what Microsoft approaches to a SQL Server and data overall, and in fact we’re doing one here in another week on business analytics and business intelligence and how this are all working in the vision going forward and where the investments are happening and where should you spend your time now and where you should do it in the future? And so we think machine learning is actually far more mature than people like to think to this. Especially in a traditional business environment, there is a lot of stuff that is out there that the community has developed over the years either on R or Python. Now is the time. If you were ignoring it before saying we got to wait and see, now is the time to jump into it. The other area that we’re telling them to get into is taking a look
(00:30:00) now at better business intelligence and data sources on premises. This whole Power BI Report Server if you guys haven’t spend time in it we’re telling them to bring that in and take a look at it. One of the other one that is kind of surprising is any traditional ETL workloads that you have going on were you’ve been suing integration services for years. You know, ETL is changing, the idea that SQL Server’s database engine has the ability to go out and tap these resources in real time and do transformations. You might not be doing ETL the way you did for the last 10 years. Integration services, I don’t see getting a whole lot of investment. I see a lot more going into analysis services. We’re going to do real time queries and being stuff together and transform it there not in Integration Services. And so I think that’s going to be kind of a change for organization. You know, and that’s going to be one if you’ve invested in SSIS you’re not going to go out and swap it next month or next year. This is going to be like a multiyear thing. New stuff we bring in, we’re going to actually bring in directly. We are not going to run it through an ETL.

Carlos: Well, now see you’ll forgive me I guess maybe that my connections aren’t quite there. How am I getting the data? When I think of SSIS, I am thinking about it is that glue that takes me from my transactional system, into my cube, or it’s doing some transformation there for me. There are additional components now. I mean, we’ve talked about Polybase, so SSAS, the analysis services, now has these connectors to actually go out and grab this information in an ample time.

Andrew: Oh yeah. 2017 baby! Here we go, so Analysis Services in 2017 supports the M language. So all of the sudden all of the stuff you we’re doing in Power Query and Excel and you’re bringing stuff in and grabbing different sources, and maybe going out and getting a file list instead of just another data source that’s all coming in 2017 Analysis Services. So the big reason for that isn’t necessarily to make analysis more functional which was kind of nice outcome from it but they wanted to bring Power BI, and Power BI is all based in Power Query. And so in order to support that with Analysis Services in the new Power BI report service thing which is basically Reporting Services plus they had to bring in the Power Query engine capabilities.

Carlos: Ok, so now there is another dot connected because I know at least the initial version of Power BI Desktop, so the new version that you can download and install locally that only supports connections to SSAS. And I guess it makes more sense now if SSAS has a bit more functionality in it for you to take advantage of some of these other data sources because I think initially when I looked at that I thought, “Oh, that’s all going to be all cube data. I don’t want to always do that.”

Andrew: Yeah exactly, and it’s not. So what is Power BI report server when you bring it down and you install it basically we work on that reporting services plus because it’s just really the same thing. There are a few other things. You’re on a modern life cycle for updates which I don’t particularly like but that’s reality. But you can host your PBIX files and so you’ve got that capability. But the PBIX files that’s the only one that’s limited right now, the rest of the RDL stuff and KPIs and all the roster are full capability. So the PBIX files that you have there can only go through an analysis services cube which actually winds up being your gateway which then you have to think of in that scenario analysis services being like SSIS. It’s the glue that goes out and grabs everything and brings it in, right? Now, that’s the first three to Power BI report server hosting so you’ll see that expanding. They are going to all kinds of data sources for PBIOX files so I don’t expect that to be limited for more than a few months.
Carlos: Ok, very good. Andrew, thank you so much for visiting with us, great conversation. I wish we could go on for a bit longer but I know that we need to wrap it up here. I guess last thoughts. Steve, do you have any last questions.

Steve: No. I think that there is just a lot going in this space and keeping on top of it is going to be interesting I think. Like we are talking with Power BI so Microsoft is investing so much in Power BI these days that everything it touches is sure to expand down the road in the near future. So I think that’s a pretty safe bet. I
(00:35:00) think there is a lot of interesting things to come here in the near future.

Andrew: Have you guys ever wanted to talk about that one in a separate one. Power BI is one of the other cover, not for this broadcast obviously but in the future.

Carlos: Yes, of course we’d love to. Again, particularly for this conversation we wanted to have somebody who was obviously familiar kind of covering that beat if you will but didn’t work for Microsoft.

Andrew: They are not even a customer of ours.

Carlos: To see just how that continues to play out. I think we touched on it a bit but we as database administrators we have our homework cut out for us. I think it will be fun to see how that continues to evolve as time goes on. Shall we go ahead and do SQL Family?

Steve: Yeah, let’s do that, so Andrew how did you first get started using SQL Server?

Andrew: Up until 2008 I was really kind of a Sybase kind of guy. I know it takes me back a while. We had a need for standalone database engines and Sybase was one of the early ones that allowed you to have a decent relational database environment that you can have on a separate machine. We have a bunch of travelers so they needed that and I did not want to go into Access or FoxPro which is where we originally started. So there’s Sybase and then as things progress we saw the direction for SQL Server and that became our standard and we moved over in 2008 and never looked back.

Steve: Ok, interesting.

Carlos: There you go. Now, as you’ve been covering SQL Server since then and then working with it. If you could change one thing about SQL Server what would it be?

Andrew: Yeah, you don’t like this. I changed Management Studio actually.

Carlos: Oh boy! Here we go. You know, because they have broken that out. They have their own development cycles now. Let’s see. What don’t you like about SQL Management Studio?

Andrew: Well, I like a lot that’s in it. I think the thing that would benefit them is as if they would make it easier for new entrance. One of the things that I find that is most difficult with new developers and new database people we bring in isn’t learning SQL or learning database structures. They’re getting all kinds of education on that. But setting it up, accessing it, using it, you know, some wizards something in there for the new guys to be able to get in and use it would be nice.

Carlos: Sure, kind of updating that almost to like a web type interface which is a little more intuitive. We’ve have all those buttons everywhere and you’re like what kind of I’m looking at.

Andrew: Yeah, and they got some stuff that’s nice but few years back we would seek new people and say, “If you really want to see how to build a query on this? Let’s do this in Access. See this nice wizard, we draw all these tables together.” Oh look at that that great query. You know, the query works on but into Management Studio and running it as a query there, “Look it works.” Why can’t they do that there?

Steve: Very good point. And I think, I mean personally I love what you can do with Management Studio but I have definitely seen the challenge when you have someone who is first time ever using it. They sort of get dumped in and don’t know where to go. There is this huge hierarchy in the preview in the left side that you can expand. How do I get to other query? Can be a little the bit daunting in the beginning?

Andrew: Oh yeah, the wizards for setting up jobs was nice. Being able to go through and take a look at especially some of the new views and performance capabilities. These are all great stuff but I got new people that we would like to get engrained in this just to make their lives a little easier so they get addicted as well.

Steve: Alright, so what is the best piece of career advice that you have received?

Andrew: Early on someone told me, “Don’t get worked up about office politics.” I said, “Look, focus on doing a good job and the people you really want to work for will notice that and promote you and those who don’t care about the good work you do.”

Carlos: Good things will come to those who wait.

Andrew: Oh yeah, and if they don’t then you’re in the wrong place.

Carlos: Sure, yeah, and then think of all the time that you don’t have to spend worrying and talking about, “Oh, what are they going to do with.” You know, I don’t like this person.

Andrew: Exactly, yeah. Just don’t worry about it. Just go to your job.

Carlos: Very good! I like it.

Steve: You know, that’s one of the things that I like about on the consulting side is
(00:40:00) you’re there to do the job and not to be in the politics. I like being on that side of it.

Carlos: Andrew, our last question for you today. If could have one superhero power what would it be and why do you want it?

Andrew: I know he gets ridiculed all the time especially on big bang but I really like Aqua Man. And his ability to breathe under water so that I could basically do scuba without any kind of limits. That would kind of eminent. Lots of people is cool too but this ability to breath underwater that’s my thing.

Steve: Hey, if you figure that out let me know. I’ll go with you.

Andrew: Yeah, but not around here. It’s too cold. We’re going to go south.

Carlos: I said that would be the last question but just one follow up. Where is your favorite place to scuba dive then?

Andrew: That’s going to be the Caribbean and basically anywhere, 7 mile off coast of Mexico, any of it. Just go down about 70 to 100 feet and forget the world exists.

Carlos: And you’re there. Wow! Andrew thanks so much for talking with us today. We really appreciate it.

Andrew: Oh, you bet guys, that’s a lot of fun.

Episode 107: BI for the Little Guy

BI for the little guyData analytics is all anyone seems to talk about anymore.  We’ve even caught the bug at SQL Data Partners and have started working with data platform MVP César Oviedo from Costa Rica.  César has been helping us on some projects and we thought it was time we introduced him to everyone.  On the Spanish front, he has been very busy putting out YouTube videos and publishing content for BI LATAM (Latin America).  We invite him on the show to introduce himself and our conversation turns to the idea of BI for the little guy–smaller organizations that need to take advantage of analytics and how the ecosystem is changing to support them.

We are happy to introduce César to you and know you will enjoy hearing from in this episode and future episodes.

 Episode Quotes

“The problem that business intelligence has had in the past is the thinking… Ok, the BI is for Enterprise. The BI is for huge companies.”

“Again a small or medium or huge company needs an architecture behind to support all your solution.”

“Another piece that is rally fancy today is data science… Everybody is trying to understand and talk about data science.”

“I think the days of the DBA who’s only responsibility is backups have come and gone.”

Listen to Learn

00:30 A little background on César
02:00 General view of today’s topic – Business Intelligence
02:52 Companero Shoutouts
03:59 Announcements about the Companero Conference
06:44 Microsoft released updated drivers for PHP to connect to new SQL Server
08:41 SQL Server 2017 – High availability support for SSRS
10:46 Why should small and medium organizations care about business intelligence?
14:14 What are the available options that you can take advantage of some business intelligence?
17:25 Tips or suggestions on PowerBI
21:12 Other things to look at when implemnting business intelligence: data warehouse, data lake etc.
23:51 About analytics and data science, and the pendulum swing of job roles
27:00 SQL Family questions

YouTube: https://www.youtube.com/bilatam
Facebook: https://www.facebook.com/BusinessIntelligenceLatam/
LinkedIn: https://www.linkedin.com/in/cesaroviedo/

About César Oviedo

César OviedoCésar Oviedo is a Data Platform MVP with more than 15 years of experience working in multidisciplinary projects running roles such as Scrum Master, Project Manager, BI Developer and DB Administrator. With experience leading teams and projects with more than 30 members. Currently focused on Data Science and Big Data platforms. Community Manager at “Business Intelligence LATAM” (BI PASS Chapter Leader for Costa Rica and PowerBI User Community), international speaker and trainer for Data and Cloud technologies.

Episode 106: Temporal Tables

User: What was that record before it was updated?
Me: I don’t know.  Maybe I could restore a backup?
User: Really? You have to do all that? It is just one record.
Me: (under my breath) Yes, a record YOU updated.

If you have ever had a conversation like this you know how difficult it can be to implement auditing of records as they change.  With temporal tables, a new feature, we have the ability to track point in time information about a record without the huge expense of setting up auditing and tracking.  While you won’t want to use this on every table, when there are tables with sensitive data you want to audit from time to time, this feature will come in handy.  We are happy to have Randolph back on the program with us to talk about this new feature.

Episode Quotes

“That ability to kind of see what was there before it was updated, that seems to be the focus of temporal tables.”

“It’s all effective dated, so every row on a temporal table has to have a start and an end timestamp.”

“You don’t know how much space it’s going to use until you start doing some actual real life testing… That’s the biggest limitation for me.”

“When you create a temporal table the history table only takes the clustering index, so mainly the primary and the structure of the table.”

Listen to Learn

03:28 Temporal Tables
05:44 Auditing relative to user tracking
08:20 What are other reasons that people putting in temporal tables?
09:49 Start time and end time parameters for temporal tables
11:39 Using system_time in querying
12:50 Recovering data
13:35 Flushing the history table
14:49 Temporal Table in Management Studio
16:17 History table and clustered index
18:01 Modifying/Altering Temporal Tables
21:17 Data consistency check dealing with corruption
22:44 Scenarios where SYSTEM_VERSIONING = ON
27:50 Other limitations of temporal table
28:39 Is system time applicable when querying joined temporal tables?
31:06 When should you use temporal tables?
32:23 Tips on current feeding side of temporal tables
43:52 SQL Family questions

About Randolph West

IndexesRandolph West solves technology problems with a focus on SQL Server and C#. He is a Microsoft Data Platform MVP who has worked with SQL Server since the late 1990s. When not consulting, he can be seen acting on the stage and screen or doing voices for independent video games.

Transcription: Temporal Tables

Carlos: Ok. Well, Randolph, welcome again to our program.

Randolph: Thank you for having me. I’m surprised I survived the last round but this is going to be interesting.

Carlos: Yes, we actually split that into two. We had you on for a double header and so it’s been about 30 episodes and so we wanted to have you back. Of course we’ve been planning with you for the upcoming conference. The Companero Conference and so we’re excited to have you speak there.

Randolph: I’m excited to speak as well. It will be very cool because the cloud is all very rara shiny but some reasons why you shouldn’t just jump into that.

Carlos: Sure. It’s interesting you bring that up, so I just happen to be reading today, so Microsoft used to be $4 billion, they said $15 billion last year in the cloud.

Steve: So we will be covering how to appropriately take advantage of that $15 billion.

Randolph: That is exactly right.

Carlos: Being milked for a big percentage of it.

Randolph: Well, some of that budget is going towards trying to convince people to go across to the clouds so it’s an interesting time.

Carlos: That’s true. But today ultimately our topic is temporal tables.

Randolph: Temporal tables, yup, one of my favorite features of 2016 and upcoming 2017.

Carlos: Right, and I think ultimately or what we’re kind of getting at here is they try to give us some additional ways to do auditing, right? They tried with some like change data capture and maybe even some other components. I think we all have a lot of the home grown components for audit but that ability to kind of see what was there before it was updated and whatnot. That seems to be the focus of temporal tables. Would you agree with that?

Randolph: I would agree up to a point because temporal tables do not keep track of the user that performed the change.

Carlos: Got you. Ok, very good. So that is kind of the distinction when we talk about the word audit, that the person itself was not captured only the value.

Randolph: Yeah, so temporal tables I’ll quote kind of what Microsoft is saying is that it is for auditing in forensics but you don’t know who the user is that made the change so you will still have to keep track of that. But it also helps, and this is where I really like it for, data that is accidentally change or accidentally deleted. Plus it also provides a point in time analysis of what the data looked like in the past, so you could say, “Show me what the table looked like at midnight three weeks ago”, and then I can see what changed during that time using trend analysis or some type of complex query and we can analyze what’s happening over time using the data that’s in the history table compared against the data that is in the current table.

Steve: Yup, and you know that’s a real important feature there I think would be on the auditing side of it because so often you’ll get a report that shows like the last 30 days sales or something like that and then once a couple of days have passed it may have change drastically and oftentimes you don’t have a good ways of seeing what that looked like last week, or the week before for comparison and temporal table seem like a great way to be able to do that.

Randolph: That’s why it’s my favorite feature of 2016 because now exactly that you can back in time and say, “Hey, this record changed.” And we can see exactly when it changed and if you are keeping track of your logins and things like that this is where the auditing comes in but also with the tracking changes in a more sensible way. There are obviously limitations to how this works because you’re eventually going to run out of space but we can get into that if you like.

Carlos: Do you think, I guess, are those features going to continue to be separate do you think and I guess we’re kind of prognostic getting a little bit here, right? Does the user going to make it to the temporal table feature or is that going to be remain outside of what this feature set is.

Randolph: I think if they want to pursue an auditing feature then they do need to include the user at some stage. Saying that though there are a lot of systems that don’t use system IDs for tracking users. They have their own. For example, when you log-in into a web application your user account may just be a table entry inside the user table so a row and a table. There is no necessarily way a good way of checking which user that is from the temporal table point of view because it’s not a system user that accessing it because there is only one system user if you like. They probably will bring it in eventually. I don’t know. I’m no using any MVP magic here. I don’t know. But they probably will bring it in but at the same time you should be checking that anyway if you have your own table of users that you’re using to check logins. I think you can combine the two and see who’s doing.

Carlos: Well, that’s an interesting point, I guess one that I haven’t considered so having not use temporal tables before. But, yeah, that idea of let’s just account has access to the application but then the application talks to the database with a single account. That architecture, that should change a bit to take advantage of some of that, right?

Randolph: Well, I don’t think you have to change the architecture. I think you just have to be aware that the users accessing the data are internal to the database or to the SQL instance and in that respect you would have to keep track of them separately. Your architecture must have to change but I don’t see temporal tables’ architecture changing if they do add system user tracking at some stage.

Carlos: What are other reasons that people putting in temporal tables?

Randolph: The big reason for me is history, keeping track of history. So one of the systems I built awhile ago was to keep track of all records changes that happen in the system. So whenever a row is deleted or updated, I keep a copy of what it’s used to look like in an auditing table and then if I ever needed to go back into the history I would have to go and query with horrific XML queries to try and figure out exactly what is going on. Because for example if you’re keeping track of all the history for all the tables, you can’t just go and create a beautiful table structure on that, you have to use XML. You’d have to shred the XML to go and read that, and that would either require an index that is 5x the size of your data or a very slow query as it goes read everything. So what this does is in a way that you would create a trigger on a table to keep track of the old row and then store it somewhere. This does exactly the same thing out of the cover. So when you create a temporal table and then you create the history table that is linked to that whenever a row is deleted or updated in the current table the old row will be kept in a history table just like you would do with a normal trigger. The functionality out of the cover is exactly the same.

Carlos: The advantage is it’s going to give you a little bit more detail. It’s going to add that begin and start date or the date parameter too. It’s going to take care of all that.

Randolph: It’s all effective dated, so every row on a temporal table has to have a start and an end timestamp. It’s datetime2, the full 8 bytes, so datetime2 goes down to 100 nanoseconds. That is the smallest granularity and that’s the granularity you have to use for system date times for the temporal tables. So there is a start time and an end time. When you create the table the first time your default value for your create time will be system date time, so current date time if you like. And then the end time will be some time in the future that you’ll never see in real life so the year 9999 December 31st for example. And then if you see a row in the history table that has an end time and then you do not find that row in your current table that means it was deleted. Otherwise if the row exists and the current table it was never deleted, it was just changed. So a little bit of trickiness to understand how things work but it’s all date time specific.

Steve: So then take the example of someone who is not quite on SQL Server 2016 yet and they want to use something like this. They could just create the same thing with their own trigger and with the similar naming convention on the start time and end time and all of that. However, where we get challenging it seems like we haven’t talked about querying it would be when you’re trying to query that data.

Randolph: Well yeah, so the query uses a different type of querying than we’re used to. There is a keyword that you have to use, system_time, so you have to use the keyword system_time in your query so, SELECT whatever from the table FOR SYSTEM_TIME, and then you can do an is at a certain timestamp. You could use a BETWEEN. There is a number of different things that you can use that will automatically figure out the timestamps are between the current table and the history table.

Steve: And to me that seems where some of the real value is and that even though it just triggers under the scenes and all that the value is the way that SQL Server has built that in to be able to query for specific point in time or range or set of values.

Randolph: Yeah, and that’s a value for me as well so I can go and see exactly what the data looked like at that specific point in time down to the 100 nanoseconds. So if we find out that somebody deleted a whole bunch of rows we can go back to the exact point in time before it happened. We don’t have to go and restore an entire database or a file group restore for doing Enterprise edition. But we don’t have to do that anymore. We can now go into the history table, recover all those rows, then flush the history table because otherwise you’re going to have 3x the data that you originally had, and I’ll explain that as well. So if you’re recovering data you go into the history table, you get those rows out, you flush the history table, and then you input the rows back in to the current table then you’ve got your consistent data before the delete happened and you reconnect the history table to the current table again.

Steve: Ok, so you use the term flush the history table there. What exactly does that do?

Randolph: Well, the history table will grow and grow and grow until you are run out of disk space. That’s by design so there is no aging out of the data. You have to do that yourself which is good because if you’re auditing you decide how long you want to keep your data for. Same that though the data is stored in a compressed formats. So they use page level compression on the history table. And because since 2016 Service Pack 1 we don’t have to worry about compression because it’s part of the product in Standard Edition as well so we can have page compression for the history table so it will use less space than it would normally which is one of the advantages of the feature on 2016.

Carlos: Now, remind me, because I guess one of the things that we’ve kind of implicitly said here is that we have a table, now we want to add this history to it. We’re going to use an ALTER Table that’s going to change and we’re going to say, “Hey, this is now the temporal table.” It’s going to create this history table. Now, we
mentioned querying it but from seeing that history table like in Management Studio, is that visible to me or is that still an object that I’m managing or is that kind of just behind the scenes.

Randolph: You can see a temporal table in Management Studio as long as you obviously have the right permissions to view them but it will show up as what’s called a system versioned temporal table because the system is versioning the table. I mean, this is Microsoft’s naming conventions between O and love. What you’ll see in Management Studio is you’ll have the current table. Let’s say for example, dbo.Account, and then in brackets behind it will say system versioned, and then if you expand that out it will show you the history table under that first. And then it will show you the columns keys constraints and all of that stuff that you’re used to. So the history table, you can name it yourself, and put it in its own schema which is what I recommend. Let’s say you have a file group that is on slower storage or cheaper storage whatever and then you put your schema attached to that file group and then you put the history stuff inside that schema then you can manage the history in a different file group which might cost you less money. It will be a little bit slower but it’s history data so you would not have it in the first place.

Carlos: You’re not expecting super fast performance.

Randolph: Exactly. And because it’s compressed you might actually offset some of that performance problem by having it compressed and then you’ll have the list of columns that would look exactly the same as they would in the current table except you cannot obviously have a clustered index. I’m going to change that statement. You can have a clustered index but you cannot have a primary key in the history table because the primary key is in the current table. So the history table does have a clustered index it just doesn’t have a primary key because the primary key is in the current table. Otherwise, the structure is identical.

Steve: And I think the key behind that is that when you’re inserting into that history table if you had the primary key on the same thing as the original table it wouldn’t allow you to put those multiple rows in there.

Randolph: Correct, so what’s happening in the history table is that you have a clustered index that is just not unique because it’s in the pinned only way of writing to the history table.

Steve: So then if we’ve got that history table and I want to clear out some old history in there can I just go in and delete data out of that table?

Randolph: You have to break the link between the current and the history table to be able to do that which isn’t complicated at all. There is just a command and if your editor will bear with me. It’s probably going to be Carlos, right?

Carlos: No, Julien.

Randolph: Hi Julien! Bear with me here.

Steve: ALTER TABLE table SET (SYSTEM_VERSIONING = OFF).

Randolph: Yes, that’s the one, Steve. So modifying temporal tables, you have to have administrative rights to be able to alter a table. You need the admin role to do that. So you cannot alter the objects until you have administrative rights so that’s the first thing. Anybody can read and write from the temporal table and write to the temporal table but only the administrator can modify it. And what you do there is: ALTER TABLE name of the table SET (SYSTEM_VERSIONING = OFF). Then you can do whatever you like inside the temporal table with the full understanding that your auditing trail is now broken.

Carlos: Sure so almost like a maintenance window type.

Randolph: Exactly. So this is exactly the same thing as changing your database from full login model to simple login model is the recovery model so now you have basically truncated or change your history so now you will have to recreate that connection once you’re done flashing the table or deleting rows from there that you need or changing column names or changing columns themselves, deleting columns, updating columns, whatever. Because you may change the structure of your current table and you want to have that change reflected in the history because unfortunately although you can do an ALTER TABLE which changes a column or drops a column on the current table that will take place on the history table. But if you have decided to keep the history of a certain column and you do an ALTER TABLE on the current it will drop that column from the history which you don’t necessarily want. Then you have to break the connection, go into the history table, SET that column to NULL, and then recreate the connection again once you drop the column from the current table. So that way you can keep history of what the column used to have in it but then the column does not exist anymore in the current table so you still got that history retained. Does that make sense or does it just confused you all?

Carlos: No, no.

Steve: I think it make sense.

Randolph: Ok, good. There is also another feature which I have never turned off. When you create the SYSTEM_VERSIONING on a table you can alter an existing table and create a temporal table out of it, which Carlos alluded to earlier, or you can create a new table from scratch which is a temporal table. And to change an existing table you have to ALTER the tables SET (SYSTEM_VERSIONING = ON) then you can assign it to history table name if you want to create something in your own schema that we spoke about. And then there is another feature called data consistency check = ON. The documentation says this should always be ON to ensure that the history table follows the correct structure format as the primary table. I’m not sure why you are going to turn that OFF. So I’m not sure why the feature is even presented as an option.

Carlos: Right, because you made a change to that table, data type change most likely or I guess you can add a column.

Randolph: But why have the history table if you’re not checking the data consistency? I’m not sure why that’s there but I’m just letting you know that it is there. It is required with the ALTER TABLE statement if you’re going to use SYSTEM_VERSIONING = ON but I’m not quite sure why it’s there.

Steve: So one thing and I haven’t use the data consistency check one way or the other but one thing I was reading about on a blog post was that it was there to help deal with corruption at some point which immediately I perked up a little bit and thought, “Well I need to understand this one.” But I haven’t had a chance to dive into that yet. Have you heard anything about using that around corruption?

Randolph: I have not but I would imagine that if you have a corrupt table, and this is all conjecture on my part. If you have a corrupt table and you have a history table that contains all the history, the history should probably be fine because it’s stored in different file group for example or different part of the file anyway. You would have access to that history without necessarily reading or writing whatever is in the current table using just standard queries. I believe that the work was a bug in one of the Cumulative Updates did fix a bug in data consistency check for certain tables, so I’m not sure if that’s what you’re referring to, Steve.

Steve: Yup, and I guess that’s something that I need to look into a little bit more.

Randolph: Because I do know that there was in Cumulative Update 1 there was a bug with data consistency check so maybe that’s what you’re referring to.

Steve: So let’s take an example here then. Let’s say we create a table and we have SYSTEM_VERSIONING = ON and we have the history table associated with that then we insert a single row into that table. Does that row then exist in the table as well as the history table at that point or does it wait to get to the history table until it ends up being changed?

Randolph: An INSERT will only affect the current table. The only time that the history table is involved is if you modify that data. So when you insert a new row into a table there is no reason for it to have changes at all so you’ll only see it in the current table and then when you modify it the timestamp will be kept track of and the original row with the start time that you created the row in and an end time of when it was changed will show up in the history table.

Steve: Ok, got it. So then if you’ve got a table then for instance it’s something like bank account transactions which is probably the type of table that’s not going to change frequently. It will have a lot of inserts but very rarely I would hope our transactions in the past being modified.

Randolph: Well, I’m going to stop here and say that’s a bad accounting practice if you’re changing because you should be posting changes in an append only model with accounting.

Steve: Sure, exactly. But if you wanted to track to see if anything was actually changing it, you could turn this on, go back and look and see what point in time things were changing if there were. But then I was going to sort of flip it around to a different example of let’s say you have a table that’s like your users table and that user has username, and a last login date where every time that user logs-in it will update the last login date. That would be one that if temporal tables were turned on every time the user log-in it would be inserting another row into that history table, and it could grow very quickly over time.

Randolph: It could, yes. And that’s why the page compression is going to be very helpful from that point of view because that data should compress very well. But saying that it may not be a good choice for a temporal table although it might be a good choice if that’s what you want to do is audit and see how often people are logging in and things like that.

Steve: Right. So then in the case that you have a very narrow table, maybe there’s just username and last login date, something like that, primary key, whatever you need there too as well but if that’s probably going to work better in a history table than if you’ve got a user table that might have a 150 user attributes to find in different columns. Again, a poorly defined table there because when it logs to the history table, of course it has login the entire contents of the row. Is that correct?

Steve: Correct, so wide tables are going to be slower because it’s exactly the same effect as if you were taking from that and using a trigger to write it to a different table. It’s still going to be a wide table, possibly wide columns as well so it’s going to take awhile to write and you are going to have a performance impact on your current table because it has to commit the row to the history table first obviously. Let me just confirm that. Actually I do a blog series in November 2015 so it’s fairly fresh in my mind. I just want to double check here. The row will be modified in the current table and then it will be modified in the history table. It’s part of the same transaction though.

Steve: Alright, so are there other limitations that people should be aware if they are going to try this out. The big thing for me is running out of space because it’s very easy if you’re not use to what kind of data or what the…

Steve: I think we’re relying on Julien a lot on this one, Carlos.

Randolph: Well, how many times did we pause the last one? That was pretty crazy.

Steve: Quite a few.

Randolph: Can you repeat the question, Steve.

Steve: Ok, what I was just going was other limitations?

Randolph: The big one for me is the disk space because you’re going to run out of space quickly. The other problem is if you are using this in a staging environment and you have more than one table with a foreign key relationship and both of them are temporal tables. You’re going to have a hard time trying to pin down history queries to say what the data looked like at this specific point in time because of those foreign key relationships so it gets a little bit hairy in that respect. You don’t necessarily want to have every table in your database a temporal table because it’s going to make your queries complicated and it’s also going to slow you down because it’s going to be querying the history over time.

Steve: Ok, so in the SELECT statement then when we were querying for temporal data where say FOR system time as of and specific date time too. Now, if you do that in SELECT FROM one table INNER JOIN another temporal table, so we have two of them there. Is that system time parameter going to apply to both of the tables and the join?

Randolph: What a fantastic question. I’m going to say that that time is specific to each table. What I would do is I would run an experiment and then find out because I don’t know.

Steve: Ok, it’s one of those new features where there is a lot of experimentation to do there I think.

Randolph: Where the challenge comes in is you may have history on a lookup table for example. So the lookup table has a whole bunch of stuff that might change once in a while and you don’t necessarily look at the history when you’re joining that to something else, but you might have to do that and it gets a little but hairy. It’s not difficult, it’s just time consuming for the engine to go back into the history table as well, and then because when you do a temporal query. I’m going to call it a temporal query and not a history query. When you do a temporal query against your table it’s going to combine the results from your history and your current. If you’re asking for everything so it’s going to do basically UNION ALL or a UNION because the data shouldn’t be the same. When you combine the two, if you’re going to do a point in time query it’s going to go and look in the history to see if anything is there so it’s always looking in that history table. So if that is on slow performing hardware then it’s going to take longer, that’s kind of where I’m getting to.

Steve: Ok, that makes sense.

Randolph: I was just going to say when should you use temporal tables? You should use temporal tables on data that needs to be tracked and that you were tracking in an interesting way before. For example, my system that was keeping track of the XML data version of history. I needed to check that for auditing reasons. That is a good reason to use temporal tables. But I wouldn’t just go turning it on everywhere because eventually you’re going to forget and eventually you’re going to run out of disk space somewhere.

Steve: Yes, and hopefully you know about that before you run out. Hopefully you have alerting in place.

Randolph: Right, you should have alerting in space but if you’re playing with it on a test environment or something and you forget about it. You may need a massive amount of space to handle that data. Some people might have to retain 7 or 10 years of history. And then you may have to age in the manual process anyway. So you don’t know how much space it’s going to use until you start doing some actual real life testing. That’s the biggest thing that I have to say. That’s the biggest limitation for me. I think I said that a hundred times.

Steve: How about any other tips on sort of current feeding side of temporal tables?

Randolph: Well, back to the disk space that’s being used. If you are creating a lot of churn and you’re temporal table is growing excessively you may want to think about going into the history, breaking the connection and deleting or aging out data that is older than a certain number of months or years. And it’s got indexes on it so you might want to do some index maintenance as well. You can apply indexes to columns inside the history table manually if you like by the way which is pretty cool. But you cannot change the clustering key so you may want to do some index maintenance from time to time on the history tables.

Steve: Ok, so if you have a history table associated with your temporal table and you add indexes, non-clustered indexes to that for instance, do they automatically end up in the history table or do you need to go and add those manually if they are needed?

Randolph: When you create a temporal table the history table only takes the clustering index, so mainly the primary and the structure of the table. They should be no reason for it to take the indexes across because that’s additional maintenance that’s not necessary because you’re not necessarily be querying that table a lot. So it will only copy the structures, so the columns themselves and whatever your primary key is will become a clustered index on the other side. There will not be any indexes that get copied across. If you do want to query that table frequently you would have to add your own indexes.

Steve: Right, so in that example if we are querying for a certain point in time and filtering on a column that needed an index, if that history table got large it could be quite slow if we don’t add that index themselves.

Randolph: Yes, and there is a whole can of worms here with filtered indexes which don’t always work the way you expect them to which we won’t get into. I think we covered that in Episode 81.

Steve: Yeah, great, great. So any other current feeding tips to add at this point?

Randolph: When you make a change to your current table the change if it is structural, so you’re deleting a column or modifying a column may affect the temporal table as well, the history table. And the reason for that is because it wants to keep a consistent check between the two. That’s the data consistency check that’s happening. So if there is data in the history table that doesn’t satisfy the requirements of the new column, for example you may have just deleted everything from the current table and then alter the column. If it is attached to the history table still and you do and alter column and it then tries to change the data type, if the history table contains a data type that is incompatible with the new change it will fail.

Steve: Oh, so you either purge the history table or know that column in the history table perhaps in order to be able to modify it in the original table.

Randolph: For me personally I would break that connection completely, rename that history table so that it’s archived and then create a new history table going forward if the data types are that significantly different because effectively it’s a new table. If you do want to go and query that history you still can by querying the history table directly in the archived version. But if you’re going to be making structural changes to your current table, I would break the connection, rename the history table and just create a new history table going forward. The other thing that I wanted to mention, we touched on very briefly earlier, is if somebody deletes, let’s say runs a delete statement against the current table without a WHERE clause so every single row gets deleted. What you will find is that the original version of that row for every single row will find its way into the history table. So if you’ve got 10 rows, that’s fine. If you’ve got 10 million rows, you’re going to have 10 million rows in your history table as well. So if you want to recover that data you can just copy that over from the history table into the current table but I do recommend breaking the connection first because any modifications you make to any rows after that will also be written back to the history table and you still got millions or 10 million changes in that history table. So either flush those rows out first or rename the table or do something but don’t try and go and copy those 10 million rows back into the current table from the history table without doing some maintenance on the history table because otherwise you’re still going to have those 10 million records there waiting in the history forever and ever.

Steve: Yup, so it really seems like it just keeps coming back to the history table can get really big depending on what you do and it’s maybe slow if you don’t index, it may run you out of disk space. I mean there may be any number of issues associated with that history table becoming gigantic.

Randolph: Well, it’s a usual story, how much are you querying it? Does that mean we’re going to be using a buffer pool for that table? All of those things that go along with that because even though it is compressed on disk whenever you read it into memory it’s going to be uncompressed.

Steve: So another question around this. Let’s say you’re querying your original table and you’re not using the system_time for system time parameter, you’re just doing a direct query of your source table and your temporal table without the history. Does that work exactly the same as it did if you did not have a temporal table?

Randolph: That’s next in question because it brings up another thing that I didn’t mention. Yes, it works exactly the same as if nothing had changed. So when you create a new temporal table you can create the start and end time using the datetime2 data type and you can mark those columns as hidden. There is another feature in 2016 for hidden columns and this is specifically for supporting this feature. So you hide those columns so that any queries that do, and you shouldn’t do those what people do a SELECT *. Those columns will not be visible in a SELECT *. They will not be visible in any existing applications that don’t know that they are there because the system is managing them. It’s as if you’ve created a column with a default value because that’s effectively what it is.

Steve: But those have to be specifically marked as hidden for them to not show up in the SELECT *.

Randolph: Correct. But if you do that and you can do that very easily when you’re doing the CREATE or the ALTER table to make it a temporal table, you can mark those columns as hidden. So it has no effect on any current applications or any queries that you might be using currently. It will work exactly the same way as it did before. So unless you’re using that system_time query keyword it’s going to be carry on reading from the current table as if nothing else has happened.

Steve: Ok, great.

Carlos: Good deal, lots of good stuff today. As we’re going to finish up here, I know we’ve done SQL Family before, but can we go ahead and hit it again?

Randolph: We can do it again because my answer may have changed.

Steve: So Randolph how did you first get started using SQL Server?

Randolph: Accidentally, as with everybody I know. Nobody I know has ever wanted to be a DBA and then they realize they’re being silly because it’s actually what they wanted to do all along. I used Access, I used MySQL, I used Ingres. I used SQL Server when I was working with PeopleSoft back in the 90’s. I didn’t know what it was. It’s just one of those things, oh this just keeps your data, it’s in an Excel
spreadsheet basically. Lots and lots of Excel spreadsheets. And then I started reading about corruption because that happened to two of the clients. The databases either got dropped or corrupted in some way and I start of getting interested in that way. And basically it’s just, it was there, it was there, it was there and let me have a look at it and then I started loving it. Access was really my entry point into SQL Server because Access was so terrible.

Steve: You know, that’s so interesting because at first glance if you don’t know databases very well, Access looks pretty darn cool.

Randolph: Access has some very nice features for front end development for a data backend. Well, it did. I mean it is 2017 now.

Steve: Sure, but back in time it did.

Carlos: Well, they have it all in one application too, right? Two things.

Randolph: Exactly, and it was easier to use in FoxPro. It was cheaper than FoxPro and you didn’t have to be a programmer to write a query because it had that cool query designer that you could do. I believe SQL Server has the same thing but I never used it. It was just really neat to be able to picture your data with a graphical interface.

Steve: I wonder if Microsoft realized that access is really a stepping stone to get to SQL Server from… I’m sure they do.

Randolph: It’s a gateway drug.

Carlos: If you could change one thing about SQL Server what would it be?

Randolph: Last time I think I said that I want to do a TempDB for every database. I still kind of think I want that but I’m not sure if that’s the one thing that I want to change now because firstly Azure SQL Database has come a long way and in a very short time so if I would change one thing. It would be making Azure Database cheaper. Because I think there is a lot of potential for people to use for single application usage. For example using an Azure Database to replace what people use MySQL for now and that would be web applications and things like that. If it were a little bit more cost effective with all the feature set that they have because everything that’s in SQL Server 2016 is in Azure SQL Database including temporal tables on the Premium Edition and in-memory OLTP and all those things. So I think temporal tables are actually supported on the Basic Version for what it’s worth.

Carlos: I think you’re right. It’s interesting because you look at the, what’s the edition that only allows you to go 10GB.

Randolph: That’s Express.

Carlos: Express, thank you.

Randolph: I have a funny story about Express off the topic. A customer of mine had a customer because they are a vendor. A customer was running their product in Express Edition. They had 20 databases and of course Express has auto close enabled by default so a typical diagnostic session which takes 5 minutes normally run for 2 hours because it had auto close on every single database. So every time a query run there we close the connection then open it again for the next query. It was fantastic. So what were you are going to say about Express Edition?

Carlos: Well, I guess I’ll say because it’s free, right? At least my impression is of making like Service Pack 1 some of those analytic features available in the Express Edition. That is a direct response to MySQL ecosystem, so I think to your point of like, “Well, hey that’s free.” Now there is compute, right, so I guess there should be some cost there. But a database under 10GB should be pretty close to free, right?

Randolph: Agreed, but…

Carlos: That might be the change there. But I don’t think a lot of people are going to take that into consideration because even if they had to pay a penny it’s still I had to pay something and this is “free”.

Randolph: Exactly. The other cost with Express Edition which I love is the 1GB limit of memory so non sense.

Carlos: Oh right, I forgot about that.

Randolph: MySQL doesn’t have that limit obviously but it is a strong consideration for not using Express Edition. You’ve only got 1GB of RAM.

Carlos: Is that still a case in 2016?

Randolph: Yes.

Carlos: Ok.

Randolph: 4 CPU cores or one socket which are very small, 1GB of RAM and 10GB database size.

Carlos: Got you.

Randolph: I would like to see a better costed version of Azure SQL Database for the Standard level if you would like. And get rid of Basic because I don’t see the point of Basic. 5 DTUs isn’t enough for anybody. I’m sure I just coined a new phrase. And then just have Standard and Premium, and have the Standard do more for less basically because I think the adoption would be higher if they did that.

Carlos: Well, they’ve been coming down. I know the sizes have been getting larger and so, yeah, and so maybe another year or so we’ll get there.

Randolph: Maybe by the time this podcast goes up.

Steve: You’ll never know. What’s the best piece of career advice that you’ve received?

Randolph: I’ve been working for so long I probably forgotten most of what I’ve heard but the one piece of advice I give to people because I do mentoring as well is to say, “Be honest”. So if you screw up tell somebody that you screwed up. It’s much easier to handle a problem if you’re completely honest and say, “This is what happened, this is how I created this problem.” And then it’s much easier to fix than lying and say, “Oh, I don’t know how that happened”, or “I did not kick the power supply out of the server”, or “No, I did not accidentally bumped this hard drive off the disk”, you know stuff like that. Just be honest and it’s just easier for people to handle the fallout. And if you’re honest it shows that you are human and there is nothing wrong with doing something wrong. I make mistakes all the time the difference is I own up to them. I’m not like, I’m not going to say it. I’m going to get political. I’m going to stop here.

Carlos: But it’s interesting on that one because I’ve heard people say that they are sometimes afraid to be honest because the environment they work in is too aggressive or management doesn’t have your back on anything.

Randolph: Well, then I wouldn’t be working in that environment. And being honest and saying, “Yes, I did this”, gets you fired. Well, then any place is better than working in an environment like that. Toxic work places are very bad for your health and you will die young.

Steve: Yup. And I think what I was going to say around that is that if you’re on that toxic work environment, and I’ve been there in the past, honesty is even better at that point. I mean, it’s always good but it’s even more important at that point because you’ll know exactly where you’ll stand.

Randolph: It’s refreshing. It might not be the outcome you expected but it is refreshing for everybody to say, “Well, this person is telling the truth and they’re not going to compromise whatever for the sake of looking good.”

Steve: Yup, very good advice.

Carlos: If you could have one superhero power, what would it be and why do you want it?

Randolph: Last time I said, I have a superpower wherever I walk into a server room a machine will either crash or not, that’s a superhero power I didn’t want. But there’s been a whole lot of superhero movies recently, a whole lot of stuff that I don’t enjoy because everybody is destroying everything. The one superhero I do like is Batman because he is not a superhero. He is just a guy. He does have compromised morals and he’s very very wealthy. I don’t want to be wealthy like him because that also corrupts you. But I do like the idea of just being a decent person. I’m not necessarily killing bad guys but if you see something going wrong, speak up say something. You don’t have to be a superhero to be Batman. So you don’t have to be a superhero to say, “That’s wrong, this is not cool and do something about it.”

Carlos: There you go.

Steve: Yes. I like that.

Carlos: Thanks so much for being on the program today.

Randolph: Thanks for having me and I’m sure this would have been a lot longer without the edits. Thank you for making me sound good.

Carlos: No, it’s great to have you on again and chat with you. And yeah, we look forward to seeing you in October.

Randolph: Sounds good. I’m looking forward to the Companero Conference in Virginia, in Norfolk, on the 4th and 5th of October, 2017. And it’s a single track so you just have to stay in the same room and listen to people explain how awesome SQL
server and the cloud is.

Steve: We’ll see you there.

Carlos: That’s right, you took the words out of my mouth.