 Live from Austin, Texas, it's theCUBE. Covering DockerCon 2017, brought to you by Docker and support from its ecosystem partners. Welcome to SiliconANGLE Media's coverage of DockerCon 2017. This is theCUBE, I'm Stu Miniman. My co-host for the next two days is Jim Covellis and happy to have as our first guest on the program is Brian Graceley. A year ago, actually, Brian had a beard and he was one of the hosts on theCUBE. He's now with Red Hat. Brian, welcome back to the program. Stu, great to be on this side of the table again. Good to see you guys. And Brian, you were at the first CUBE event back in 2010. We've had you on at least once or twice every year. You did a few more when you were on our team, but happy to have you back as a guest. Why don't you bring our audience up to speed? What brought you to Red Hat and what's your role there and what brings you to DockerCon? Yeah, so been at Red Hat about a year, a little less than a year now. Worked on the OpenShift team, so focused on Kubernetes, containers, integrating Linux. It was a great opportunity to be in OpenSource, which I've been working on for a year. It was at home, it was in Raleigh, and it's a great team. It's a team that's growing. The Kubernetes space is growing, so the vendor side of the world drew me back into Red Hat, so it's been good. Yeah, OpenSource, big component about what we're talking here at the show. I heard OpenSource mentioned it done. It was developers, it was contributors. What's your take? Did you get a chance to see some of the keynote? Solomon got out there, thanked the 3,300 plus contributor when he put up the name of companies. I think it was 41% of the contributors for all of this are independent, but then Red Hat's in the top six companies there. What's your take on that and the ecosystem in general? Yeah, I thought the keynote was good. Obviously, the show's doing well, so it's great to see the container space doing really well. We've been part of the Docker ecosystem since sort of day one. We like to say that we're probably the biggest distributor of what used to be Docker is now Moby within Raleigh. But yeah, I think we see that, we obviously believe in the OpenSource movements. We're seeing more and more customers, our customers who want to contribute, who want to make it the de facto buying decision as to what they do. So yeah, it's great to see not only huge OpenSource support, but then seeing it become, to blossom into very viable commercial offerings around the market. Yeah, so Brian, your team actually wrote a blog leading up to shows as containers are Linux. After listening to the keynote with Linux kit announced, it felt like, oh well, Linux is containers. It seems like, remind me back, Sun is the network is the computer, the computer is the network. It's all kind of looking at it. What's your take as to kind of the relationship of containers with Linux, of course Windows fits in the mix too, but the operating system and the containers? Well, I think, the reason we really put that out was, if you go back a little bit historically, not to bore people, containers aren't a Docker thing. Containers are a Linux thing. They were created by Google, Red Hat made a huge contribution sort of secondarily around namespaces. Google did C groups, IBM did LXC, so it's been a core Linux feature for over a decade now. Docker did a great job of making it easier to use, but at the end of the day, even if you look at what Linux kit and some of these other things are, they're not about sort of Linux versus Windows or they're all Linux and it's, how do I represent Linux in ways of doing that? So we really kind of want to just reinforce this idea that there are things that you expect out of your operating system, containers being one of them, but if you look at every other project that's being built around this space, whether it's Kubernetes, whether it's the management tools, they're all being built on Linux. That's the foundation of this and it's kind of just a reinforcement to people that remember where your tools come from, what that thing is that drives security for you, things in that space. Brian, you wrote a lot about kind of cloud native and that journey kind of rewriting applications, containers, fits into that a lot. What have you seen changing kind of last 12 to 18 months? Couple shows I've been to lately, it feels like we're talking about lift and shift more than we are about building new applications. What's the application space look like? And I know Jim's going to want to jump in here. He covers the cloud native stuff. So I think there's a couple of big things that, and I wrote about it for a while and it's how much it's changed in the last two years has been really interesting. So I think originally when you went and looked at platforms, whether it was OpenShift or Cloud Foundry or Heroku or whatever, lots of sort of what we used to call opinionated systems. You dictated what developers did, right? And then we had... Kind of opinionated systems? Very opinionated platforms, right? The opinion of us, the creators was going to get forced on you, the developer, right? Yeah, it's made a lot of the decisions for you. Right, and again, the idea was make it easier for you. You don't have to think about those things but you're going to get them in the way that we want them. And what ended up happening was, Docker kind of became a standard. We had a standard container format. We ended up having these open source schedulers like Mesos and Kubernetes and other things. And that allowed the platforms to be a little more what I was calling composable. So because developers may not want to use the languages that you force on them. They may not want to use them in those ways. So I think what we've seen is this sort of blurring between what used to be heavily opinionated to becoming more composable and modular. And there's always this trade off between how much do developers want to care? How much do they not? So that's one big trend that we've seen is this sort of back and forth of what that is. The other one we saw was... It seems a compatibility, but I'm going to go quickly. Do you see any trend in this space containers for visual composition of applications? What I'm seeing in today, and I've seen generally in this space is mostly coding, command line interfaces, any visual composition tools you guys provide or any partners of yours for building containerized applications? So I think there's sort of two pieces that it's a great question because ultimately, if the coding piece is hard, you only reach a small segment of those developers, right? You want to, it's like when websites came out, they were all hand coded in HTML and stuff. And then you had things like Dreamweaver and these other visual tools and it exploded. We've seen that to be successful in this, you've got to have tools on the desktop that make it easier for the developers. Red Hat does something that we call the container developer kit, which is really write your application. A lot of the stuff in the background gets hidden. Docker has Docker for Mac and Docker for Windows. We see some other tools. So that piece is important. The other piece that to come back to your question about is it lift and shift? We probably see 75, 80% of the customers we work with who say, look, I know I've got to do cloud native. I've got digital transformation and all these sort of things, but I've got a lot of portfolio that I'd like to modernize. Can I do that with containers? And I think what we've seen is for the early days it was containers are only for new. They only work for microservices. They're only for new. And what we're seeing, and this again goes back to the sort of containers are Linux. As customers say, I have an application that ran perfectly fine in Linux. Why wouldn't it run really well on there? And we've got customers nowadays and this sort of blows people's mind. Like we've got customers who will pick up things like WebSphere, put them into a container, run them, modernize them somewhat, but because the platform will give them automation, it gives them high availability, it gives them scalability and they go, it works and they get cost-effectiveness. So we're seeing a lot of that because you can address a lot of your portfolio. Brian, it's the typical maturation that we see. The use cases they've put on stage, keep planes in the air, power the largest infrastructures, monitor fire alarms, websites. Oh, this is the same thing we saw in virtualization in every kind of wave. It's like, oh, containers run applications. Right, right. Is there a big push by your customers or in the ecosystem to containerize more of the deep learning and artificial intelligence toolkits like TensorFlow or Theano? Is that, with your customers, is that a big priority right now or going forward? Yeah, so I think the big data space was always an area that was kind of on the fence of it made sense in containers. Do you need an abstraction layer? Do you want to be closer to it? We're starting to see more and more. So for example, Google with TensorFlow, Google, huge proponent of containers and Kubernetes, they're doing a lot of work to make that happen. We've been doing a lot of work with the Spark community to make Spark work really well in containers and it becomes an issue of, can you manage the resources? The container schedulers do that great and then can you manage getting access to the data? And we're seeing more and more storage become container native and people understanding how that works. So yeah, the breadth of what you can do around containers has gotten very, very large. Any difference in how your customers look at it, whether they're doing on-premises or public cloud or do things like Docker and Kubernetes make that not matter as much? I think what they, so, you know, I joke all the time like, none of our customers have a container problem, none of them wake up in the morning and say that's my problem, right? What they're saying more and more is, I know I want to, I'd like to start getting away from maybe owning data centers or my destiny being data centers. I need to leverage public clouds, multiple, plural. And they're sort of saying, look, I get the benefit of what they do but there's still operational differences. What Azure does, what AWS does, I would like some level of consistency. And so that's where the OpenShift conversation really comes into play. The operational model I can build with OpenShift as a platform is the same thing I can run on top of Azure, on top of AWS, on top of Google. And we're seeing more and more of our deals, our customers who say that's what it's going to look like, help us make that work. And today they do it, you know, on a basic level, you know, somebody like Volvo, for example, some in their data center, some in AWS. And then more and more, they go, go contribute upstream in Kubernetes and federate this stuff, make it look more consistent, make it look more operationally consistent. And that's coming in like the next version of Kubernetes and so forth. So that shift is happening, but what they want is sort of this consistency. The Kubernetes part, the Docker part, they're sort of details under the covers. But it does provide them a level of portability that's really important. All right, Brian, want to give you the final word. Red Hat, you know, we've got Red Hat coming up, OpenStack, I know Jim Whitehurst is going to be giving I think the day one keynote there. Right. Talk a little bit about Red Hat's presence, you know, here at the show, what we can expect to see in this space from Red Hat throughout the year. Yeah, so I think from us here and what you'll see at Red Hat Summit, like containers are front and center. You know, obviously it's an extension of Linux, but it's, we're becoming a company that's more about how to do applications faster, how to modernize applications, how to do them across multiple clouds. And it's this whole idea that those things that used to be really hard, you do them in software now and the community is helping to fix those. So, you know, big presence here, you know, again, we've got a ton of customers who use Docker as a packaging format, run their containers, open at Red Hat Summit, we're going to have 25 plus production OpenShift customers that, you know, you want to talk about running governments, running airplanes, running, like, they're going to talk about that stuff. So that part we're really excited about, it's fun, it's fun at this point, they don't, our customers don't want to talk about containers, they want to talk about this digital transformation stuff. Making the technology industry fun again. All right, so that was my last question for Brian Gracely with Red Hat. My last question for Brian Gracely at the Cloudcast is, I haven't heard serverless mention yet this week, what's wrong? I know, that's a good question. The serverless stuff's taking off two weeks from now, probably at the same, no, down the street, serverless comp is happening. Is that part of OSCOM then? No, it's its own event now. The serverless comp events are their own event, they'll probably get 500, 600 people. We're seeing it as another way of looking at applications. Functions, containerize them, write your own code, and you'll see us, you'll see what we're doing around OpenShift begin to incorporate that sort of functions as a serverless stuff very, very soon and around Boston time frame. All right, well Brian, always great to talk to you and glad I can bring it to the audience. So, Brian Gracely with Red Hat, we'll be back with lots more coverage here at DockerCon 2017. You're watching theCUBE.