 Live from Vancouver, Canada, it's theCUBE at OpenStack Summit Vancouver 2015. Brought to you by headline sponsors EMC and jointly by Red Hat and Cisco with additional sponsorship by Brocade and HP. And now your hosts, John Furrier and Stu Miniman. Okay, welcome back everyone. We are live here in Vancouver, British Columbia for Silicon Angles theCUBE, our flagship program. Go out to the events and extract the signal and noise. We're live at the OpenStack Summit, extracting all the data, sharing that with you. I'm John Furrier with my co-host, Stu Miniman. And our next guest is Tim Eton, SVP of Infrastructure Business at Red Hat. He runs all development, essentially, one of the head honchos. Welcome back to theCUBE, great to see you. Thanks for having me back guys, good to see you too. You get Red Hat Summit coming up in Boston this year, not San Francisco, so on the left coast. A lot of action happening. We're here at OpenStack. What's the development action? I mean, it's in your world. Red Hat's used to great support big customers. OpenStack's developing very, very fast. Absolutely. And it's on a path to maturity, not yet truly mature yet. What's happening? How are you guys viewing the market and the relationships you're building? Well, I think the first thing is we do agree with the point of view that it's maturing rapidly. Even as we're driving a lot of innovation, if you look at the next release of OpenStack, you know, a huge set of technologies focused on the telcospace with NFE. So the pace of innovation isn't slowing down, but at the same time, I think we see the core getting more stable and supportable. And for companies like us, you know, we understand that the community model is always going to be about innovation. So, you know, no one in the upstream is going to build a production quality product. And so we gear everything that we do to, you know, taking the innovation as it comes, the more robust the better. But we're going to put it through our products, let's say, regardless for, you know, making that admission critical, consumable for customers. Look at the quality of the releases. Are you seeing better quality coming out of the releases? Are you happy with it? What's your take on it? I mean, you guys have a high standard, given the tier one position you have in the market. But I mean, this is emerging, but getting better, what's your take on it? Better? Do you get a good vibe on that? So definitely better. And as you noticed over the last year, so we've extended our support life. So for any given letter release, we're not providing three years of support. So a customer who takes this into production isn't forced to then, again, every six months update, the minute they can stay on for three years and ultimately will continue to extend that. And it's very dependent on both our confidence and the robustness of the core and our ability to make it bulletproof and supported. So we're already at three years and we're intending to extend that as we go forward. So definitely getting there. And I think if you just look at the buzz here, the attendance is great. People's confidence is high and you're hearing lots of stories of production customers now. I mean, there's people on the floors. I mean, the sessions, just for the folks watching, I got just to share with you that if you're not here, it's packed. There's technical sessions. I mean, literally people are on the floor with their laptops open. I mean, it's like school. It's like a builder community. A real transition quickly from the thought leadership in the early days to now, you know, engineers are here, building and wanting more data. Well, I can just, just from, you know, Red Hat, where I think we're hosting 44 sessions. And that's sort of as requested, you know, and they're all technical sessions are all how-to kinds of things, you know, hints and kinks. So, you know, the people that are here are clearly on an implementation path, which, you know, even if you think back a year ago to Atlanta, there was still a lot of, you know, is it ready for primetime yet? And I think we're through those, and now it's a question of for what use cases, you know, at what scale, the kinds of good questions that people should get. Yeah, primetime, no, basically. Yeah, exactly. Because primetime just started. Right, exactly. So Tim, you know, obviously Red Hat has a huge presence at the show. I've seen lots of Red Hats being handed out, people walking around with them. One of the things you're highlighting at the show is your partnership with Cisco. Can you talk about, you know, why the partnership with Cisco and how you're helping together to help customers modernize their partnership? Sure, absolutely. Well, you know, Cisco and Red Hat have had a long-standing partnership, you know, in the UCS space. And, you know, a year, year and a half ago or so, we started to recognize that we had a joint and shared vision around the evolution of cloud being open, being hybrid. And we started on a path to collaborate, you know, in CCS with Fiasas Organization, as well as other parts. And the great thing is, you know, they've got, you know, tremendous presence in the networking space. They work with a lot of customers. Frankly, that, you know, we've served as, you know, data center customers, but I think this is, together, given us the opportunity to help reshape how, you know, telco and networking infrastructure is playing out, leveraging open source, RELL OSP. So we've got a very multifaceted set of activities and, you know, the companies, you know, have huge numbers of engineers dedicated to building in all these areas and making RELL OSP within Cisco solutions real. Yeah, one of the key things that was talked about this week in the keynote was, you know, interoperability and portability. Maybe talk a little bit about, you know, how Red Hat and Cisco, the solutions you're doing, kind of tie into those. Well, I think, you know, it's great to see the foundation, you know, championing things like, you know, the federated identity, you know, now you're going to get interoperability at a user definition level across clouds. You know, I think just the very nature of OpenStack is such that you can build cloud-native workloads in a very consistent fashion. Cisco, for example, with their intercloud strategy, you know, is consciously trying to create ways to allow clouds and workloads on those clouds to interoperate both those that they're hosting and those that we're jointly partnered with. So we're happy to be working with them, being an enabler for that, but I think in the end, you know, we're going to see these dense fabrics of hybrid clouds, you know, public and private, and we've geared our whole product strategy both directly in with Cisco to ultimately respond to that. Right, how about if going up the stack into the application stays, how does OpenShift fit into this discussion? Yeah, so great question, and in fact, we made some announcements this week that we can touch on if you like, that sort of allude to this, but, you know, OpenShift is our Paz platform, but, you know, I think that actually does it a bit of a disservice because it's really the culmination of all the things that we're doing to enable people to both modernize traditional applications and have them live in a cloud-native environment where OpenShift winds up being the pinnacle of that, where it's highly opinionated, but also highly automated, but all of the things that we build OpenShift out of are embodied in our other platform technology. So as you look at OpenShift version three, it supports, we'll support, you know, Docker as a native container format. Kubernetes for orchestration, it's based on our relatomic host, which is our specific container footprint for rel. It's all managed by cloud forms, our cloud management platform, so you get a continuity of whether you're deploying Paz style with OpenShift or IaaS style with, you know, OpenStack, and now we have OpenStack as a way to scale out your OpenShift environments on-prem, so we've built this really nice sort of fluid environment where the choices you're making aren't Paz versus IaaS, and whatever development models predominate there. It's, you know, what services do you want to be? Highly automated, preconfigured, and then just give me a way to manage, you know, build applications across them and manage them across them. I like this vision, I want to expand on that a little bit, because this gets down to what the language of the customer, no one talks, you know, infrastructure service, because there's now a service architecture, so you guys have a unique vision, I want you to share that. Absolutely. I think I've made an announcement, congratulations on that, but I want you to tie it together, because we are living essentially in a service-oriented architecture. Web services now, all that stuff from 10 years ago is actually on the main stage in production. So what does the service architecture look like and what does that mean in the language of the customer? No one really cares about platform as a service or softwares, it's all one big cloud rollout, so I want resources. And in fact, old timers like us look at, you know, microservices, our architectures say, isn't that so? You know, we've seen this before. We've seen this before. Well, and I think that goes to the point, right? So, our view is, you know, there are classes of applications that, you know, are either stateful, you know, scale-up, database-centric, and most of those still remain very strategic to those customers, right? They just don't lend themselves architecturally to a scale-out, stateless kind of model. And then there are, of course, what we always talk about in terms of cloud-native apps. Those can be, you know, also enabled by VMs or someday soon via pure containers. So our view is, you've got classes of workloads that actually don't need to be stove-piped to an infrastructure, so take containers as an ultimate evolution. You know, you can containerize elements of an existing, you know, traditional app, you know, to make maybe captive data available by wrapping it, and have that be consumable by your cloud-native stateless app. So our view is, you know, VMs versus containers, that's not the discussion. It's what do the workloads require, both in terms of architectural approach, what things they're trying to consume, and have those live fluidly in that environment. So it's runtime assembly of resources, not so much, you know, this platform, and I've got all, you know, general purpose, whatever, locked in. That's not, that's what people in the industry put things in boxes, but what you're saying is, customers are thinking like a programmer, and they're saying, okay, I'm assembling resources, and I have to program it, so I'm really talking to services. So, okay, I get that. But I got to ask you, what the hell does microservices mean? We hear that at the Cloud Foundry Summit. Is it a buzzword? Is it specific in meaning? Is it just a service of a service? Is it just, I mean, what is a microservice? It's actually a very simple context. And I'll put it in a specific sort of container dimension. So today you can containerize an app, single app, you know, one container host OS. That's interesting because you do that a lot of times, you can get much greater density. But where it gets really interesting is when you start to build multi-container applications that may or may not live on a single host, right? So multi-container single host, multi-container multi-host. Well, the value of that is you get, you know, the runtime assembly kinds of elements so you get much greater reuse of components. Serviceability and support, people haven't even thought of, right? So if I have security vulnerability and I've got, you know, my registry of containers, I know what applications are consuming, what containers, and I build workflows that just update and automate that in a really elegant way, rather than trying to find, you know, 300 instances of an app that have an SSL bug. Or you're doing an audit, basically. You're deep in the weeds on, you know, going through data. Right, it's all automated, right? You just build the workflows and all that becomes automated in the container. So you can literally contain threats, meaning the container, but like, your efficiency isn't off the charts. So what you basically get with microservices is you get these loosely coupled service-based applications where, you know, I'm coding a lot less, I'm consuming more of what's existing. And I can also give assurance and I know where those things came from, what's in them, and. To the customer, what does microservices mean? I'm like, I hear people talk about it. It smells like a buzzword, it smells like hype. What is it? It's an assembly of multiple containers in an app where you're probably inventing a whole lot less and just organizing and orchestrating. Yeah, so Tim, you know, microservices are great. You know, one of the things I've heard from some people is just like everything else, you get a little bit of sprawl. So, people create those services just like we had VM sprawl, we can have some of the microservices sprawl. You can, but I think microservices gives you the opportunity if you build up a registry of certified containers, you now can pull from those whereas every VM, you know, was sort of constructed specific to the application of the workload. So, you get much greater reuse and supportability. So, you can propagate these changes just by automated workflows. Whereas VMs, you know, you had to go fix every one of them. All right, so, we don't have much time left. Can you talk a little bit about just kind of the customers that you're showcasing here at the show? Where are our customers? So, the good news, as I mentioned, is we've got now hundreds of paying customers around OpenStack. What's really interesting is if you look at those, you know, about half of those consume OpenStack and our CloudForms Cloud Management. So, to us, that's a really good sign because in the end, you need to manage the environment and the workloads on OpenStack. So, you know, customers are already starting to conclude that. I think, you know, an interesting sort of example of a customer that really gets it is FICO. If you look at what they've done, so this notion of, you know, Paz and I as sort of fluidly built together and you're just making service selections, that's fundamentally what they do. They happen to use all that infrastructure to do that and CloudForms to manage it. But, you know, I think we're seeing that pattern more and more and more. The other thing that you'll see here is, you know, the customers and partners that we have. We talked about Cisco, but, you know, we've got numbers of partners here, you know, that maybe in the telco space, you know, people that work with like the Nokia's, the Huawei's, the Alcatel Lucent's, and I'll forget your 20 others. But, you know, and what's exciting is they're all sort of seeing the reality of it's a really heterogeneous hybrid cloud world and they know that it's more than just the stateless workloads, it's how you bring it all together. Well, I think, I want to say congratulations. I think your vision's amazing on that. I agree with 100%. I think the service model is right on. It's layers and containers are underneath that. They're all out there, but then that's not the definition. It's not how customers think. Certainly you can put things in boxes, but like, come on. You know, John, you're a really important point that I think even at a forum like this, sometimes it gets lost. You know, depending on which study you believe, you know, the majority of the world's workloads won't be cloud-enabled for a while. People are moving there rapidly, but in the end, you know, there's a lot of vendors that talk about, you know, just the net new workloads and what lives on the public and private side and refer to that as hybrid, but we see it as much broader than that. How you bring your traditional apps that aren't going to change into that world and leverage what's there and make it all work. And to roll it out in an SLA, I mean, you start with three years, but I know your goal is to get 10, but I mean, three years is still good, considering what's how new it is. So congratulations. Thanks. I really appreciate you coming on theCUBE, sharing SVP at Red Hat. Breaking it down, I always said you guys have a good systems background. We're in a systems programming world. Microservice says it's all happening. Stu, we're breaking it down. More after this short break, we'll be back live here at Opus Tech Summit. I'm John Furrier with Stu Miniman. We'll be right back. Thanks guys.