Welcome back to the SQL Trail! Have you ever wanted to stick a bunch of SQL Server professionals in a room and hear their honest take on things? I found myself in Columbus Ohio at the SQLSaturday and I wanted to do a ‘live’ podcast and this episode is the result. I chat with a panel of SQL Server professionals including Jonathan Stewart, Kevin Feasel, Mike Fal, and Microsoft employee Scott Klein. The panel members chat about their favorite Microsoft features. We also discuss the “Microsoft paradigm shift” and how the hybrid model is likely to stick around for a long time.
We Talk About…
- Direct seating, availability groups, and Windows Server 2016
- Why SSRS Mobile is a must-have for any company with C-level executives
- The benefits of mobile SSRS reports for database and network administrators
- The paradigm shift that’s spreading across all Microsoft products
- PowerShell changes coming in SQL Server 2016
- The problem with the “cloud first” model and why hybrid is the future
- The power of Query Store
About Kevin Feasel
Kevin Feasel is a Database Engineer at ChannelAdvisor and a Data Platform MVP out of Durham, North Carolina. He was also a contributor to “Tribal SQL”, a DBA guide from Redgate. Follow him on Twitter @feaselkl.
About Jonathan Stewart
Jonathan Stewart is the Principal Consultant at SQLLocks LLC and regularly presents on BI topics. Follow him on Twitter @SQLLocks.
About Scott Klein
Scot Klein is a Senior Technical Evangelist at Microsoft and former Microsoft MVP. Follow him on Twitter @SQLScott.
MSDN: What’s New in SQL Server 2016
Windows Server 2016 Preview
MSDN: PolyBase Guide
Intro to PolyBase in 2016: Part 1
PolyBase in SQL Server 2016 (by Scott Klein)
Video: Query Store in SQL Server 2016 (by Scott Klein)
SQL Server Query Store
The SQL Server 2016 Query Store: Overview and Architecture
Security and flexibility with SQL Server 2016’s hybrid cloud solutions
Carlos: So companeros, we have quite the panel today. We’re here in Columbus live in front of a live studio audience! We’re gonna be talking a little bit about some 2016 features, so because we are going to be doing a panel I’ll let each of them introduce themselves. There’s a couple of them that you’ll have heard before. In fact, you know, Kevin man, we’re going to have to start paying this guy to be on the show. But let’s let them introduce themselves and we’ll go from there.
Kevin: Hey, I’m Kevin Feasel. I’m a Data Platform MVP and a database engineer out of Durham, North Carolina.
Mike: Hi I’m Mike Fal. I’m a SQL Server consultant with UpSearch and yeah, I’m here and happy to be talking about SQL Server 2016.
Jonathan: Hi I’m Jonathan Stewart and I’m a Business Intelligence consultant and I am a guest of these guys and happy to be here
Carlos: Awesome, so one thing I did note that you added to your resume since we last spoke, Kevin, is the MVP. It’s nice, congratulations on that. So again, ultimately our topic today is SQL Server 2016. Who wants to start? What do we want to start talking about? Here we go.
Kevin: Okay, so there are a few features in 2016 that I love. But above all, number one feature? PolyBase. I love this concept that I have a Hadoop cluster, I want to be able to access this Hadoop cluster from SQL Server. Most people in this space are going to be a lot more familiar with writing T-SQL statements than if I throw a scala application at them. They’re not going to want to maintain that. They’re not going to want to maintain dozens of pages of Java code when really all they want to do is, do a SELECT statement, pull this data from the Hadoop cluster, and tie it, join it to data that’s in SQL Server and return a results set. I want them to see this as just another table as opposed to a whole different data system.
Carlos: Yeah, that brings up a great point and Polybase is something that we haven’t talked about here on this show. But that idea of as different data sources continue to change, the landscape is changing there. In fact, that was the last podcast episode that you were on. Just talking about some of those different technologies we’re now required to support those. So having something that can kind of tie them together a little bit easier so I can use my tried and true interface to be able to do that. That’s going to be exciting. So we’ll actually commit to having another episode on Polybase and what that will do for you and we’ll talk about that at a later time.
Availability Groups and Direct Seating
So Mike, do you have a favorite feature of 2016?
Mike: Gosh, 2016. Where to begin? Um, I will say that the thing I’d like to lead off with is, I’ve been working with SQL Server since 7.0 and the 2016 release is probably the most excited I’ve been about a SQL Server release in that time. There’s just so many great changes, so many new editions. Um, my favorite edition is probably something that’s a little obscure to people, particularly, it’s around availability groups. And Availability Groups are starting to get more adoption, especially since Microsoft is adding more and more to it. There’s a lot of improvements to availability groups in SQL Server 2016, especially in combination with Windows Server 2016, which I just got confirmation is going to get released during Ignite later this year. So keep an eye out for that. But there’s this thing called “direct seating”. And if anybody’s worked with availability groups if you and to stand up a second node you had to go through this restore process with restoring fulls and restoring logs. The great thing about direct seating is that you create the availability group and the database, you say, “My nodes are enabled for direct seating” and SQL Server will populate it for you. You don’t have to do anything else and it just freakin’ works. I love it, it’s great, and there’s still some clunkiness to it. It’s still obviously in its first iteration. And I’ve got a blog post on it in June’s T-SQL Tuesday if people want to check out mikefal.net and read up on it. Um, but yeah it’s a great feature and I’m really excited about it.
Carlos: Now do you have to use the Server 2016 in order to use that feature or is it native to the database?
Mike: It’s native to the database. The Server 2016 stuff actually doesn’t have anything to do with direct seating, it’s more how you configure the cluster underneath so you can do stuff like cloud and you can do distributed availability groups. There’s a whole lot of enhancements, like you said. And server 2016 can blend into those, but that’s obviously beyond the scope of this little conversation.
Carlos: Sure, and touching base on that clustering for a second, one of the things I heard was that Windows 2016 is that now I’m gonna be able to join nodes of different operating systems and I can operate in a lower mode without having to upgrade everything. So that’ll be kind of interesting.
Analytics, SSRS, and Native Data Connectors
Carlos: So we’ve been focused a little bit more on the engine side. What about on the analytics side? Anything, Jonathan, that you’re interested about?
Jonathan: There’s quite a few things. One of the big things, which is more of an annoyance, is the ability for integration services to be able to support multiple versions. That’s a huge, huge benefit, because being a consultant and having to keep installs 2008 R2, 2005 because there’s clients on that. I got to the point where I had to start spinning up VMs because I couldn’t keep stepping on other features and stuff like that. Another thing that I’m really happy about tis the addition to reporting services, like the DataZen being built in. Because one of the thing that a lot of my clients, a lot of my clients are a Microsoft shop and so they want to stay on Microsoft products. You know, and it’s a lot better for me to tell them, “hey, you can do this with reporting services,” instead of having to tell them that they have to go out and get one of those other services. And we won’t say Tableau, right? You know, I use it too al ot and it’s great at what it does. But for costs versus features, Micrsooft is quickly catching up to the Cognos of the market. They do big, big, big things. It’s not 100 percent there, but they’re catching up fast. So I’m actually really happy to see, and too the things that Kevin was talking about. Polybase and you know, he’s going to have a tlak later on R. R integrations. There’s a lot of stuff in the data world that 2016 has in it that I’m excited to see. And I’m going to show my age here – I started with SQL Server when I was 19. Literally. But I started using SQL Server when it was Sybase. [laugh from the audience]
Yeah, see, so that’s taking it back, back, back, right? Yeah so I remember SQL Server 6…6.5, right? 7 was like, wow, it’s a whole different game. So I’m just, [laughs] I’m just so, the massive analytical push that Microsoft is bringing in. The new connectors, the new native connectors… all these native connectors are great, because it sucked having to use connections to all these things and hoping that they support it and support that command. So having all these connections to other high end systems is great.
Carlos: And we haven’t even talked about JSON, right? That’s another one in there. Now we’re going to have Jessica Moss on the show later, depending on how these episodes get laid out, one of the things we’re going to be talking about there is mobile. Give us, if you will, if you can, give us like the 30 second elevator pitch for SSRS Mobile.
Jonathan: One of the great things about it too is that you’re dealing with CEOs and other C-level people who need to make decisions on the fly when they’re in meetings, and they don’t have time to you know, open up their laptops. And we always make fun of CEOs when I see pictures and stuff like that, but there’s actually legit reasons for that, because they don’t have time to open up a workbook and go through multiple excel spreadsheets just to see how much money they’re spending, blah blah blah when they’re trying to pitch their stuff to another CEO. So opening up an iPad and be able to break out and show a graph or something like that on the fly, that’s one reason that the Tableaus and stuff like that of the world were so great at what they did. Speed would definitely add to that.
Kevin: So Jonathan was talking about C-levels and higher ups. This also makes sense for the administrators. Okay, I’m not always in front of my computer. Sometimes an alert happens and I want to check out what’s going on in my system. I have a report that says, “These are nodes that up. These are the nodes that are down. Here are systems that are running, here are how they’re functioning.” If I have an ETL process that I’m concerned about, what’s the status of that ETL process? Make those mobile reports and make it so that I can see, do I need to log into a computer and check something out?
Carlos: Okay, so knuckle-dragging Neanderthal that I am, when I hear mobile reports, let me give you a bit of context. My world of SSRS, one of the things you don’t need is IIS to install SSRS. Now all of a sudden what I’m hearing, which makes sense, “Oh, I can see it on my mobile device.” Okay, that kind of makes sense too, I don’t know why you couldn’t have taken those reports and made them miniature? So now, is there a requirement to have SSIS? So I’m now exposing these out on the internet, is it really just as simple as that?
Jonathan: There’s an application that you install whether you’re using an Android or you’re using an Apple. I would hope there’s a Microsoft application for the three people who use Windows phones. And [laughing], I guess they’re out of luck, they have to get real phones. But there’s always Bing. But there’s applications, though, and you port it to the application and publish there. But one of the great things about it is it’s not just a regular SSRS report that’s miniaturized. It’s actually designed for the mobile format. So when you’re designing in visual studio, you can decide how you want your layouts to be. So think of a visual dashboard and how this lays out. You can design with things like that. So it’s meant to be extremely visual so you can process the data really quickly and go about your day. So that’s one of the benefits.
Carlos: Okay, very cool. So I think my biggest feature or what I’m excited most about, and again I’m a little more on the engine side. It would be the in-memory OLTP options, right? So I’ve drunk that Kool-Aid. We’ve had Jos, Jos de Bruijn on the show talking about that and I think that’s going to be very, very cool. In fact, he makes the argument that in certain instances you can actually use that in-memory technology and reduce your reliance on the cache. It’s like a  cache as a result. Of course networking and other things still play into that, but we’re going to see the ability to scale, right? To be able to handle more batches at one time. It’s something that I’m very, very interested in. Ultimate that we’re all interested in our environments performing a little bit better. So if we can manage that, that would be good.
The SQL Server Module
Mike: I do want to get one other little thing in there. This is like a “one A”. and 8t’s actually not quite SQL Server 2016, it’s SSMS 2016. For most people who know me, they know I’m a PowerShell guy. And I’ve had to deal, and I get it. I talk to many, many people out there and the bad rap that SQL PS has, and it’s mostly because Microsoft threw something out and it’s had a lot of neglect over the years. So, a change that’s happened with Microsoft, and this started back in April, is that the SQL server tools team has been revamping the SQL PS module. They’ve been putting in some bug fixes, they’ve been tweaking the functionality of it, to basically fix some of these problems. They’ve got a long way to go, but one of the really exciting things is here in July, just two weeks ago, the SQL Server tools team dropped an update to SSMS that included a new module for SQL Server.
It’s called the “SQL Server Module” as opposed to SQL PS, and it has all of the functionality of SQL PS but they added 25-ish new commandlets. A lot on Always Encrypted and a couple for just getting stuff like getting the SQL error log. And really the exciting thing here is it’s showing the Microsoft and SQL Tools team their commitment to, because they’ve been haring this stuff from the customer base, from the user base. Like, “what can we do to make this better, this is such a pain.”
And they’re now responding to that feedback and they’re responding pretty quickly. It’s only been three months that they’ve really been doing this work and suddenly we’ve got a new module with new command lines and there’s more coming. There’s a big community effort right now to try and drive that. So that’s my sort-of SQL 2016, but I really wanted to get on that.
Carlos: [audio cuts out]
Mike: That’s one of the things that we’re talking about, is, and it’s in process. The current release of the SSMS 2016 has this new module in it. It’s already out and people can download it and use it as part of the SSMS toolkit. But there’s more coming, and that’s the thing, there’s so much going on and it’s going on so rapidly.
Microsoft’s Paradigm Shift
Jonathan: So to follow up on your “One A” thing, and this is a “One A, roman numeral one”, that’s it, yeah. So it’s not technically a 2016 thing but it’s a paradigm shift pretty much for Microsoft. And it’s their response and the ability to make changes on the fly. I mean, just Management Studio alone that it can update independently of everything else on the fly. You know, you can check how fast they’re making PowerBI updates and changes, so not having to wait. And I think that goes back to Satya’s, his background with the games and like that too. And we’ve seen that with all these things, so if you have an update for Xbox you just roll it out. You don’t wait until the second Xbox. You change, ready, go. So instead of waiting, I’m ready now. So I was telling Kevin earlier, I want to see what 2018’s gonna bring, right? And we’re just getting to 2016, but I mean, because, what can you do to shock the world when you’re already giving us these updates every two to three months? If you’re giving us these little updates you’re obviously saving something big for 2018. So, you know, I’m not trying to throw that out to do that yet, but…
Mike: Actually, I don’t know if they’re saving something big for 2018. If you think about it, Microsoft and this paradigm shift which I totally agree with you, they’re going into this mode of faster and faster updates. You know, we already, bringing up SSMS as an example, SSMS is now on a monthly release cycle. They’re going to have updates to it incrementally. Incremental stuff, small stuff, but every month we’re going to have a new update to it. I’m pretty sure, and I’m not an MVP like some people so I don’t have the inside track on this, but SQL Server and all Microsoft products are moving in that. So when you say, “Oh, what’s going to be big in 2018?” It may not be. It may just be, “oh, here’s another incremental update that has this cool new feature.”
Kevin: In response to Mike, “NDA, NDA, NDA, NDA.” Actually, there’s also the Azure side where we’re already seeing, we saw version 12 of Azure come out pre-2016 and it had a lot of interesting things that made their way to the on-premises version. Well, Azure is going to be the playground for what the next features will be. We’re going to see them in Azure first, we’re going to see them in preview, in that public preview, before we get it on premises.
Carlos: And I think it is kind of a testament to that cloud-first, mobile-first methodology kind of trickling down. And oh! To speak of the devil, we just had a Microsoftee walk in the door. Scott, do you want to say hello?
Scott: Hey everybody! Scott Klein, Microsoft, happy to be here.
Carlos: If our conversation changes a little bit, ladies and gentlemen. That’s why.
From panel: Azure! Azure! Azure! Azure!
Jonathan: Now that there’s a Microsoft person here, I 100 percent love Microsoft. I have three tattoos. I have a tattoo that changes from SQL version to version to version. I have it
From panel: The Kool-Aid is very, very real.
Business Intelligence and Azure: Some Concerns
Jonathan: Right, right. Across my back I have Hekaton written in kanji. [laughing] I mean, and this is still on the 2016 theme. As Kevin was talking about Azure, I’m interested to see what changes are coming to Azure too, because I do have clients that want to run the current version of Azure on their database and their data warehouse. But there’s still holes, there’s still huge holes, you know like if I have an OLTP in SQL database and I have a SQL data warehouse, how do I get the data between them? Do I bring up an Azure VM to run SSIS, you know? What do I do with analysis services? I see data factory more as biz store. That’s how I, you know. I see data factory more as biz store.
Carlos: That’s not far from the truth.
Jonathan: It’s biz store for me. And what do they do with cubes?
From panel: We have a Microsoft guy here, maybe he can answer all those.
Scott: I don’t know if I can answer a cubes question.
Jonathan: ‘Cause right now, I tell them for that either they keep it on premises and they move data back and forth or we grab another VM which defeats the whole point kind of, because you have to pay for the VM and put SQL Server on it.
Scott: Or just spin up an Azure SQL VM. I don’t know.
Carlos: So I guess the question is, what is the interaction between my on premise and my Azure services?
Jonathan: The full lifecycle from your OLTP all the way to analysis cube. You can’t do the whole thing in Azure without having to step out and bring up VMs as well.
Carlos: I guess I will say having talked to Matt Usher, in that we actually brought up that point. So when I thought about Azure SQL data warehouse, he wasn’t talking about cubes necessarily. He was talking about processing, right? And so the destination is not so much the feature they’re offering as it is the computing power behind that.
Scott: I kind of get the feeling, and don’t take this as gospel, that they’re trying to shift away and sort of change, it’s a paradigm shift, right? You go from on premises to the cloud there’s a paradigm shift. And I get the feeling, and this isn’t Microsoft speaking, but I kind of get the feeling that we’re going to do BI and analysis services and kind of this different way in the cloud, right? And we may be going away from cubes or something like that. I have no idea, right? So don’t take this as gospel. But maybe we’ll see something like this down the road. But I think what they’re going with is, “How can we make this a little bit easier in the cloud?” That’s why you see Azure Data Factory and the ability to look at that Azure Data Factory in more holistic, not just SSIS perspective, right? Because you cannot compare Azure Data Factory and SSIS, I mean those are completely two different things. So I think, I’m looking at Azure Data Factory as a way to move that data and if you look at Azure Data Warehouse holistically, how do we do these things like data warehousing in a much different way?
Carlos: And I think part of that shift is that when I think of Azure data services I think of those streaming components. Much more than, again, like a cube is the final resting place. I don’t meant to say that it’s a graveyard, but
Jonathan: It’s a terminator, yes.
Carlos: I think the services that they’re building that, like, “hey I have this inflow of data and I want to see 15 minute increments,” if you will, that’s the scope of the data I want to look at. So those are the problems that I’m trying to solve because they require higher performance, they shift up and down. I don’t need this necessarily all the time. It kind of works in that cloud “let me help you expand your infrastructure in small trunks” and not have to make such a huge investment. Mike’s over here smiling.
Mike: I don’t think on prem’s the answer for this. I think the minute you have to take data and say, “I need to go on-prem to do this,” then I think you need to relook at your architecture, right? If you need that, put it in an Azure VM with SQL but I don’t think bringing it on-prem is the answer
Kevin: I kind of want to respond to that in that I agree with it on a fundamental level but I also think, we’re in a phase right now with moving to the cloud and cloud development where we still, there’s so much stuff still on prem and a lot of companies are saying, “well I can’t just pick up my stuff and move it to the cloud and I can’t just jump from one side of the cliff or the other. I have to stand on both sides of the cliff.” So there’s a lot of, when you start talking about these architectures and what-not, you really have to keep in mind the hybrid model.
Mike: And I won’t argue that, absolutely won’t argue that.
Kevin: And I think that’s kind of when you start talking about the lifecycle, you have to think, how am I going to migrate stuff up, how am I going to have both sides communicate with one another? And that’s really where I think you kind of need to be able to step outside. I don’t know, there’s a lot of stuff around it but in general as I talk and this is the point I wanted to stress is as people are looking at the cloud and they’re thinking about the services and all the offerings, you gotta stay focused on the hybrid model because personally in my experience it’s impossible to just say, “Well, we’re going to move to the cloud!” You’ve got to be able to be partway.
Carlos: And I think Microsoft has acknowledged that a little bit in their, like, “oh, hybrid! It’s a thing.”
Scott: And hybrid is just, I think, hybrid is just massively generic, right? Because there’s so many aspects of hybrid and I could be completely wrong –
Carlos: That’s why we invited you, Scott, to be completely wrong.
Scott: Completely wrong. If I’m doing BI in the cloud, I’ve made a decision to go to the cloud I think. Now there might be some outliers around that, but if I’m doing BI in the cloud I’ve pretty much moved my stuff to the cloud. I think, what does hybrid mean? I think that answer is different for everybody. Hybrid could be, “I’m just doing stretch database.”
Scott: Or hybrid could be, “I’m backing up to the cloud.” That could be hybrid. What does hybrid really mean? If I’m doing BI or I’m doing SQL Data Warehouse, I’ve pretty much made the decision to –
Jonathan: But a data warehouse is still just the storage equation. That’s just the beginning part of full blown analytics. Like, Microsoft pushed tabular for like four years, but where do you put your tabular models? You need to build up a VM to run analysis services again. Al l those things that are there that people are using today, you can’t just drop those off and go to the cloud. There’s that part of hybrid again. To me, it’s an incomplete model right now. I don’t know what’s coming tomorrow because I’m not an MVP.
Carlos: And things do change all the time, right?
Jonathan: And I heard rumors about things coming. And I can’t say who’s telling them, because who told me may be obvious if I say rumors and things that Microsoft does understand the pain that I’m talking about and they’re working to address the Analysis Services hole, because that’s a huge hole.
Kevin: For the record, I don’t tell Jonathan anything.
[laughing from panel]
Jonathan: It was not Kevin. Kevin has not ruined his 16 days of MVP status for me. Now next month…
Carlos: No promises, right? So kind of wrapping up, Scott you kind of joined us late, but our topic was on favorite 2016 features.
Scott: Oh man.
Carlos: So you’re not technically a SQL guy per se…
Scott: Oh no, I am.
Carlos: Oh you are? Okay.
Scott: I was a SQL Server MVP and then an Azure MVP before joining Microsoft, so.
Jonathan: Do you like how he changed the subject thoug?
Scott: Oh, I did no subject change here.
Carlos: So why don’t you tell us? What is your favorite feature about 2016?
Scott: Oh man. Okay, uh that’s a tough one but I’m going to say, just ‘cause you could stick with core SQL Server stuff but I think with the way your industry is going that I’m a fan of some of the big data things that are in SQL Server like PolyBase and R…
Carlos: That’s what Kevin mentioned.
Scott: Oh really? Yeah, not that… I’m very interested to see where that goes. Yeah, stretch database is awesome and I think the whole hybrid is very needed and we’ll see where that is gaining traction, but I’d like to see where PolyBase goes just because of the whole how can we use the tools? How can we leverage the knowledge and the tools that we already have?
Carlos: In a more familiar environment…
Scott: In a more familiar environment let me do this non-SQL related stuff? And that just fascinates me.
Kevin: So Carlos, what’s your favorite feature?
Carlos: So I did talk about the in-memory OLTP.
From panel – That’s not in 2016.
Carlos: Yes, well I think it’s finally put the big boy pants on in a sense. I feel like I can use it without having to baby it so much.
Jonathan: In 2014 wasn’t it a beta product?
From panel: Oh…. [laughing]
Carlos: [laughing]Scott: Kind of like, did he want MVP status? Hmm…. Let’s see, taking names!
Jonathan: I’m a Linux MVP.
From panel: Ohhh man…. [laughing]
Carlos: It’s getting deep over here.
From panel: Those exist…
From panel: When SQL Server is on Linux GA’s maybe…
SQL Server Management Studio
Carlos: So my “One A, roman numeral a” is not really a feature, but the ability to install SQL Server Management Studio without having to go through the whole install, that’s really cool. It’s kind of like, “Yes! Finally!” Yes, thank you!
Jonathan: But I’ve seen that first screen that you have to click through anyway to get to the real one. It’s like, “Oh my god!”
Mike: Baby, baby steps man. We’ll get there!
Kevin: I do want to throw out one more feature since we didn’t have many administrators in the room, unless Mike counts as an administrator. He counts more as a PowerShell guy. Query Store is a gigantic feature in SQL Server 2016. We’ve already taken advantage of it and I’m not even a Database Administrator and I’ve taken advantage of it. We’ve already identified plans that have performed poorly for whatever reason and it’s like, “Oh look! 95 percent of the time your query’s performing great because it’s using this plan. 5 percent of the time it’s horrible because this separate plan.” It’s great being able to see that. Being able to even force that plan like, “Hey, I know the 95 percent plan is really going to work.” I don’t know why you’re getting the 5 percent here. Maybe it’s outdated statistics and maybe it’s something else. But I know we want to stick it in this area. So Query Store is huge.
Scott: I agree with that one. Of all the features that should be used? That one. That one.
Carlos: Yeah, I know when I spoke with the team pre-release of course they were very excited about it. They were seeing good traction in the early release of the usage.
Scott: I think there needs to be some guidance around using that too. What is that saying? With great power comes great responsibility? I think people can shoot themselves in the foot a little bit with that.
Jonathan: Even with me in the BI spectrum, like my background where I came from I was a DBA. And even know when I talk to other BI professionals, and I’m still performance, performance, performance. And that’s a great tool. Are my queries I’m using to export data or whatever, how are they running? And stuff like that. Doing small things like that. I think just to think everybody, not just the administrators, you know Kevin said he’s not an administrator he’s just a F# guy.
(Cross talk among panel members.)
Jonathan: Yeah, that’s that other language. Yeah, it should be used by everybody. Everybody can benefit.
Kevin: I just want to say that Jonathan said that everybody can benefit from F#.
Carlos: For the record. We have it recorded everybody. Well great, thanks for the conversation. From our time perspective I think we need to wrap up, but ultimately I think we can all agree that we’re all looking forward to 2016. Well, I guess it’s here now. We just need a few more implementations of it. Thanks for coming in and being on the panel.