 OpenStack Summit 2016, brought to you by the OpenStack Foundation and headline sponsors, Red Hat and Cisco. Now here are your hosts, Stu Miniman and Brian Graceley. Welcome back to OpenStack Summit 2016, here in Austin, Texas. This is theCUBE, happy to have back on the program. Chris Wright, who is the VP and Chief Technologist with the Red Hat. Chris, thanks so much for coming on the program again. Yeah, you're back, glad to be here. All right, so day one here at the show, big keynote, 7,500 people here, lot of excitement, talking about the maturity of what's going on, talking about, you know, more production environments, talking about some, you know, solidification of some of the key use cases. So what are some of your, you know, main takeaways from the keynote this morning? Well, just that, I mean, we're really seeing OpenStack move into the next wave of maturity and that's what's so exciting. We're seeing it go from, you know, I kind of call it early stage, almost technology experiment to the POCs and into real production, where we're seeing companies, like today we announced customers Verizon and NASA bringing OpenStack to production, solving real business problems. So it's not just about technology, it's about actually, you know, what are we doing to harness that technology and change how these companies are running their businesses? A lot of things come in full circle here. The first OpenStack summit was here in Austin. And you talk about NASA, I mean, originally it was like, you know, it was Rackspace and NASA, those were the ones. Talk a bit about, you know, how Red Hat's fitting here. I mean, I look at the logos behind you on the wall there, you know, Red Hat's on top. Of course, we're happy to have a lot of Red Hat people on our program here. So, you know, how's that matured from Red Hat standpoint? Well, I mean, we've grown ourselves. So we've gone from just putting our toe in the water initially to being a significant part of the community. And from a product point of view, we have hundreds of customers. So we've gone from, it really is just a technology experiment to a real production product. And so what we're trying to do is help mature OpenStack, not just in terms of bringing it to production, but in terms of bringing it to production across a wide set of use cases. I really believe that for OpenStack's long-term success, we need to see the community bringing, you know, collaborating to allow a lot of different use cases, some of which five years ago, when we got involved initially, weren't use cases we were thinking about at all? We're thinking about private cloud or large-scale public cloud with a really specific meaning. Today, we're talking about use cases spreading across different industry verticals. Interesting one would be NFV. So yeah, it's interesting, Chris. I mean, from the early days, I've always thought of, you know, Red Hat community always went well together. What I've seen over the last couple of years that has even matured more, talk about some of the partnerships. So, you know, we just had somebody on from Microsoft from talking about Red Hat, you know, sitting in a juror. You know, talking about the announcements on NFV, you had some partners involved in this. Can you speak a little bit to how, you know, where Red Hat sits and, you know, where they need to reach out to some of the partners in the community to help put the solutions together? Yeah, well, Red Hat's always had a kind of a partner-centric, broad ecosystem view of the world. Because in order for us to be successful, we can build the core platform and our key technologies, but we're never gonna be able to satisfy all the use cases without working together with partners who have different areas of expertise. OpenStack's no different. And in the OpenStack context, you know, we're partnering with, if you look at OpenStack as a set of components, some of those components themselves are pluggable. So we have partners that, like in the networking case, Neutron, we have quite a few partners that help us build out the software-defined networking story for OpenStack, similar for storage. And it's, for the whole industry, for this project to be successful, it has to have a broad reach and a big ecosystem. And for us, that's partner-centric. Yeah, it's interesting, you know, we were talking to some people just prior to this and kind of giving them a lay on the land to new people to the community. And they said, you know, I'm trying to understand all of, there's lots of different companies here. And we said, you know, four or five years ago, there's a lot of companies saying, hey, we want to be the Red Hat of OpenStack. It's sort of come around that the Red Hat of OpenStack is Red Hat. How much of an advantage does that give you, you know, the Red Hat stack, you know, beyond Linux, obviously, but there's the OpenStack pieces of it. There's OpenShift. There's now a lot of different ways to consume it. How much, you know, when you're talking to Verizon, you're talking to NASA, people that make 10-year commitments to their technology, what does that mean to them to say, like, I'm working with somebody who thinks about things as big as I do, as opposed to, you know, some little piece? It's huge. I mean, it's a big part of the value that we bring to the customers. It's not just, you know, we're not focused on one layer of the stack. We bring in not just products, but a real depth of engineering expertise across all layers of the stack. And what we're seeing is these things are not, you know, while they're layers, they're not living in isolation. So we need to make modifications up and down the stack to really realize kind of an end-to-end solution. And our, you know, our customers are looking at, OpenStack is just a piece of the puzzle. It's infrastructure. You made that point today. It's, you know, it's sort of got its place, but it needs help with a lot of other things. Yeah, they care about storage. I mean, we're talking about significant, you know, major rollouts of storage. We're talking applications that need to sit in whether it's middleware or whether it's next generation infrastructure, building cloud native applications. These are all things that OpenStack is the foundation for and enabling, but from a full stack point of view, you know, it's actually what we've been talking about internally a little bit with some of my colleagues is it's kind of nice when we can get to the point where OpenStack actually isn't part of the story. It's a critical component, but we're starting to really focus not on just the infrastructure, the core technology components, but the real problems we're trying to solve. And those are application-layered problems. I love that. I actually, I made the analogy my sum up from last year is Linux took a long time to go from, you know, something that, you know, was really useful, but, you know, it's ubiquitous. I mean, it's in every part of the infrastructure today. And I feel like just some of the dialogue around where OpenStack should be. It's not, you know, let's rip out your entire stack and put OpenStack in there. I've talked to some software companies that when you dig into what they're doing, it's like, oh, look, there's OpenStack under the covers and it's doing some really cool stuff there. So, you know, the discussion of it as an integration engine, the discussion of where it can fit in, you know, is in some ways, you know, an extension of what had gone on in Linux in the past. Yep, I totally agree. I was thinking the exact same thing in terms of analogy is Linux is ubiquitous for OpenStack to be successful. We need to see that kind of ubiquity at that infrastructure orchestration layer. And to achieve that, if you look at where Linux was focused 10 years ago, SMP scaling, maybe 15 years ago, SMP scaling. Today, it's on your phone, it's in your car, it's on your TV, it's all these places that we weren't even thinking about, you know, when we were focused on scale up to big iron. And similar for OpenStack, we want to make sure from, you know, from a technology point of view that we're building this infrastructure in ways that we're solving a variety of use cases and, you know, some things that really stick out to me are in the Telco use case or even managed services, where you may have your control nodes for OpenStack pretty geographically separate from the compute nodes. Across a WAN, that immediately creates a new strain on the infrastructure on the OpenStack project. You know, if we can't solve those problems when somebody else can, that's, you know, that means we're missing an opportunity where we could be in that industry. Chris, in previous years, it was like, all right, what's the big component that needs work? It was like, okay, you know, Nova got knocked down, you know, Swift and Cinder, you know, it seemed in good shape. Neutron took a little bit longer, but you know, seems we're getting their adoptions up quite a bit. Are we past some of the main, you know, big blocks and, you know, it's now just feature development or, you know, where's the big effort? You know, what kind of, you know, things do we still need to knock down going forward? It's going to be use case specific. So in some use cases, it's still important to get a lot of features, functionality, or maybe robustness directly into Nova and Neutron. But in other cases, we're looking at, what does it mean to, you know, it's not day zero anymore. We're not just getting a POC going. It's actual operations, run time, upgrades. And so, you know, we saw in the OpenStack user survey and we hear from our customers a lot of interest in improving the robustness of the platform, making it easier to deploy and making it easier to upgrade because you're not going to just deploy once and walk away. This is something you're going to have in your data center ongoing. So maybe if you can unpack for us a little bit, kind of that telco NFV story. So, you know, you talked about Verizon, you know, as a customer AT&T, was also a keynote, Ericsson's here. You know, you've got, you know, some big companies that, you know, talk about reliability and uptime. I mean, you know, it can't be, you know, 99%. It's, you know, four or five more nines than that to keep things up and running. So, you know, maybe you can help us unpack that. Well, it's the kind of reliability, availability, and serviceability are the kind of core features for carrier-grade technologies. And we've looked at it from the point of view of we're not trying to create for NFV some special variant of OpenStack. We're trying to ensure that OpenStack itself is carrier-grade and can satisfy these use cases for the carriers. You know, availability is measured in different ways. Historically, it starts, you know, in the carrier world, it starts at a hardware level. So, you have redundant hardware, you have a bunch of sophisticated software managing low-level failovers, and you have applications that are kind of blissfully ignorant that all that's happening under the hood. In a large-scale data center, we've learned from companies like Google and Facebook and Twitter running massive infrastructure that components fail. And we need to help this industry move forward to define availability at the service level, not just at the infrastructure level. And that's something that we're really focused on is how do we, you know, we don't want to marginalize the NFV operator community by saying we're not, we don't care about your use cases, we don't care about your reliability and availability. We want to really help them see how we can build next-generation applications that take advantage of the fact that they're scaled out and failure has a different meaning. You can manage the failure at the application level. So, you know, everybody's got one of these these days, right? We're talking about Verizon, we're talking about Swisscom, and like that whole telco NFV space. What happens when, you know, people, like give us an example of the types of things that are changing when this becomes the most mission-critical application for almost every employee, the way you interact with, like how does that change? What is NFV doing? What are those big kind of like, we don't do this anymore, we have to think about this. Is there anything kind of immediately relatable that you can say this is a problem that Verizon had to solve that they wouldn't have had five years ago that you're now looking at? I mean, the main thing, it's, I'd say maybe less the actual applications and more of the fact that people, and so this isn't in the business context, this is in the consumer context, more of the fact that people are using the network to download so much video content. So there's just a huge strain on the network and in order to scale up the network to satisfy that demand, with a bunch of bespoke, you know, special purpose physical hardware boxes, it just is like the economics, the math doesn't work out. So the first phase is just satisfy that massive growing network demand with something that's infrastructure that's more flexible, it's made out of software, commoditized hardware. The next phase is trying to understand, okay, now that we have, you know, the phone has a first hop into the network. What kind of applications can you build when you have a great compute power right at the edge? So you've got a bunch of intelligence in your phone, you now have the capacity to put compute power right at the edge. You know, we're looking at things like, we already have self-driving cars and there's research going into augmented and virtual reality and things that require low latency because you've got a short hop to that first point into the network. You know, we're kind of just tip of the iceberg scratching into what NFE can mean as we build out this next generation network. Yeah, it has fantastic ramifications. I mean, you start, you know, the architecture is sort of always ebb and flow. It's centralized, it's decentralized, it's centralized. For a while, the telcos were scared to death about the idea of sort of the dumb internet, the dumb pipes. And now, even if they're first use cases like sort of auto-scaling, dynamic scalability, that means I watch my sporting events more enjoyably, I communicate with my family more, you know, like all those things are the things they're now selling based upon. You know, are you available? Are we the fastest downloads? That's totally different than like, you know, when it was wired connected or now they're going to be relied on for cars and for wind turbines and those things. It's fantastic to see what that begins to start to unlock and what it means from a technology perspective. Yeah, I was at Mobile World Congress over the winter and that really summed up into two buzzwords, 5G and IoT. 5G is the next generation network. IoT is an application, you know, a way to connect sensors and smart devices across the network doing new things, whether it's kind of machine to machine transactions or enabling kind of new technologies for consumers or business. It's, you know, that's, we're building infrastructure. We don't 100% know how it'll be used but we know that the ramifications are significant. Right, right. No, it's fantastic. So, you know, in your role as CTO, obviously, you're up on stage in the keynote. You've got to be able to sort of project the vision but behind the scenes, you're trying to look out three years, five years. How does that work, especially when you think about what you do with your Red Hat hat on and what you do in the community? Can you give folks a sense of how you think about what the future might look like and then how do you think about, you know, steering Red Hat resources and community resources? Well, for one thing, when you're looking out on the time horizon like that, the Red Hat and the community hat are kind of much more blurred because we don't have products built out of those next generation technologies so we're really technology focused. I think another thing that's unique and interesting in that context is historically companies would, you know, they're confined by their own resources. So they're investing into the future just based on their own ideas. What we're doing is looking around the industry where open source software is enabling tons of innovation all across the board and so part of what we do is get engaged in these new emerging communities, kick the tires, understand what it means, look at what the ramifications might be for our existing products or how it, you know, might take us to a completely new place. So it's, you know, the Red Hat, the community hat, one of the things I love about being in a technology role is that line is much more blurry and it's, you know, we're as blindsided as anybody else when something pops up on GitHub that just goes viral. And, you know, the best we can do is monitor and watch and get involved quickly and see how we can, you know, work with those innovative projects. Yeah, so Chris, one of the interesting things we've seen, especially in the OpenStack communities, Red Hat's been an acquirer of a number of technologies. The thing I looked at is I'd look at the OpenStack survey and be like, oh, okay, let me look at some of the things there. Oh, look, okay, you know, they kind of take the CentOSP, so, you know, Enovance comes in for some of it, you know, number of companies. Talk about, you know, how Red Hat looks at those acquisitions in kind of the interaction with the community. Oh, well, community is always front and center in these acquisitions, so we are looking, you know, first of all, what is the business justification? Why are we bringing this company in? Is it a technology tuck in? Is it a services augmentation? You know, what are we trying to achieve from a Red Hat business goals point of view? Any technology that comes with a company is something that we're going to take an open source. So, you know, how are we going to build communities around that? How are we going to sort of intersect other existing communities with this technology? Those are all important parts of the conversation when we're bringing companies in. And it's a way that we can really grow, you know, our products and our skill sets internally because the world's moving quickly and as much as we can hire and go within our own resources, go in new directions, bringing in new companies is another way to help us. Yeah, I love, there was a great, you know, parable, I guess, or not parable, but a description by Jim Whitehurst in his book talking about there was one acquisition that was like, ah, we'll keep it kind of close source for a little bit, but then move it to open source eventually and the end of the story was we will never do that again. It will always start day one open source and we'll do that. I guess the last question I have for you is, you know, Red Hat's much more than Linux these days. You know, how long have you been with the company now? About 10, 10 and a half years. Can you just give us a little bit of insight as to, you know, how things have changed inside for yourself, kind of the breadth of what's going on inside Red Hat? Well, it's huge, it's top to bottom, it's a full stack, it's low level operating system infrastructure, it's, you know, infrastructure orchestration, it's application middleware, it's application orchestration, it's management, you know, all of these components. You're a cloud provider in some cases and, you know, a little bit of everything. We provide OpenShift as a service online. Yeah, so we've got this broad product portfolio. We're trying to address all of our customer's needs. What it means internally is, you know, there's, I'll roll back in time, we brought in JBoss, a big consumer of JBoss technology, we're Windows developers. The Linux people at Red Hat, like, this is oil and water at this point in time, right? And today, we're seeing how all these products are complementary, we do have to keep an eye on how we're developing different components so we can see how we can put them together to build comprehensive solutions. But it's, you know, it makes for an exciting time and we're really focused today on these integrated solutions and what that means from a product engineering point of view, how we bring teams together and sort of share ideas and share information. Yeah, all right, well, Chris, thank you so much for joining us. We'll actually be happy on a little bit later in the show, digging a little bit more on the solutions that you're doing in partnership with Cisco. So Chris Wright, chief technologist, Brian gave him a portion to CTO with Red Hat. And thanks so much for joining us. We'll be right back with our next guest after this quick break. It's always fun to come back to theCUBE.