 Live from the Fairmont Hotel in San Jose, California, it's The Cube at Big Data SV 2015. Okay, welcome back everyone. We are here live in Silicon Valley for the last day of three days of wall-to-wall coverage of The Cube. Big Data SV, our event, Big Data Silicon Valley in conjunction with Stratoconference and Hadoop World. We're The Cube. We're here to extract the seeds from the noise. I'm John Furrier with SiliconANG. Our next guest is John Kleiman, who's the director of business strategy at Red Hat. Welcome to The Cube. The Red Hat Summit last year was very interesting because we actually interviewed the CEO of Docker at the time, no, actually the founder of Docker before they changed their name, I think it was. But now Docker's been the rage. OpenStack's been amazing. You guys have been leaders in that. Big Data's happening. What's going on with you guys here at the show and what are you guys announcing? What's going on with some of the activity shared with the folks, your presence here at SV? Sure. So I guess I'll start where you mentioned on OpenStack. So we've heard a lot about the cloud. I think everyone has. I was found the Salesforce keynote. Very interesting. And everyone was here, right? Microsoft's here, Google's here. And so a lot of activity on the public cloud side, but also on the private cloud side. So just earlier this week Red Hat shipped the Red Hat OpenStack platform version six, which includes Sahara, which is kind of the infrastructure that allows customers to build effectively their own elastic map reduce inside their cloud. So we're seeing a lot of interest in that now that it's out and available. We've talked about it on theCUBE before. So that's very exciting. You know, we're going to be in Vancouver this year, so we're definitely going to be up there at CUBE again to continue this coverage. Sorry to interrupt. Go ahead. Oh no, that's okay. We love those summits, they're really awesome. Yeah, yeah. So the OpenStack community is coming together and starting to embrace the big data as a workload. I think Sahara will go a long way to helping that. And then, you know, the other thing is I think a lot, you know, continuing to hear lots about real time, continuing to hear lots about analytics, and you know, basically people want to get value out of the data, so it's all about the analytics. And so in that area, we've got a couple things. One is we announced a partnership with Continuum, which is one of the biggest players in the Python space, Continuum Analytics. And so we have a, as customers are looking from scale out to scale, or from scale up to scale out in terms of their Python implementations, they need an infrastructure to put that on. So they've, they just announced Anaconda, which is sort of their cluster management product, and then we put in a connector and verified that on top of Red Hat Storage Gluster. And so we announced that, and that's up on our blog post as well as some sample code. And then obviously whenever you're doing real time analytics and trying to get value out of your data, you want to get all your data sources together. So we also announced a new Impala connector for Cloudera Enterprise for our JBoss Data Virtualization product, which allows customers to pull data out of, out of Cloudera through Impala, and combine it with non-Hadoop data sources so they can do their analytics at a higher level. So let's take a step back, because I really want, I love talking with you guys. Red Hat is one of my favorite companies, I've been following you guys from Gen One when Open Source was, when you guys really came onto the scene. By the way, it's not a Tier One offering in the enterprise. Now you're Tier One, Open Source has exploded. Look at where we're living in an era of amazing Open Source goodness. I mean, it's really fantastic if you've been a developer and you've seen where it's come from. It's super, right? So it's great. You guys are so awesome at Open Source. What do you think about the Open Source situation right now? You got so many good things happening. I was asked to comment on the open data platform that Pivotal announced with Horton, where there's a variety of other players, some big names, and I usually would be a naysayer because consortiums tend to not get under my skin a little bit, but I was critical of Open Stack when it became kind of a marketing fluff thing, and then they tweaked their mandate, ship with code, let your code do the talking, and then also boom, Open Stack's successful. Cloud Foundry, I was very skeptical on when they just came on, there's no way that's ever going to happen. But they've cobbled together a little foundation. It's working that's generating revenue. So I'm going to reserve my judgment on this one because Open Source is so proliferated now. Is it changing at all? I mean, is there a modernization happening in Open Source that's adding to the goodness of where it's come from? I mean, is there a new dimension if the Internet of Things includes machines and code and people, and it's instrumented in real time, doing things in the open is 100% open. So if it's 100% open, does that change the dynamics of hiding behind the curtain in the old days? We all know the tricks, hide behind the curtain, lock and spec, market spec. So is it different now? Are you seeing any tweaks to this Open Source model in a good way, or bad way, or nothing? Is it the same? Can you share your perspective or observations there? Because that's something that we've been talking about all week. Yeah, so yeah, obviously, we're the purist, right? So to us, it's all about showing up with code and contributing to the community, and it's a meritocracy, so the more you work, the harder you work into it, the more open you are, the better. So that's, like you said, sort of the- I think that's like, I don't think that's pure, I think that's just the way it is. I mean, code rules, right? Yeah, yeah, and that's the way we look at it. But as you said, right, there's definitely been over the last, I don't know, five to 10 years, sort of this leakage, right? There's a lot of companies, they've been successful. They all say they're doing open source. They don't quite follow that model, the Red Hat model, right, which is upstream, always first, pull down later, harden it, for the enterprise, that's your value add, not necessarily putting in proprietary hooks. So there's, but as you said, you can't argue with their success. So maybe it is, maybe it's, there's multiple models now, right, there's the purist model, which is the one we sort of live by, but then there's this, you know, happy medium or slight adjustment to it. Well, it's not proven yet. So again, I'm using all my judgment. Even OpusTac has had some growing pains, but people are finding their swim lanes, as Jeff Kelly said. So we'll reserve judgment, certainly the purist way works, right? But you got to look at that and say, hmm, this open core, dancing around the edge, you know, hard and top, where does that go to? But you know, interesting, I want to get your perspective, because we're watching this, because you know, one thesis we have is, maybe there's a new modern tweak to open source that could take it another level. We don't know, we're just watching it. So appreciate your perspective. Now back to OpusTac and some of these other opportunities. OpusTac is interesting. So where do you guys see OpusTac intersecting with big data and the data fabric, the data lakes, the data oceans that are around us now is a lot of dynamic conditions and data are changing. Real time is a huge focus. Analytics are still the game. That's the app side. And OpusTac is all about infrastructure for developers, right? So app developers. So where's this all converging? What's the pressure points that are most active in OpusTac that you guys are involved in? Yep, yep. So I mean, the one easy one is Sahara. But beyond that, right? If you think about what you just said, the two worlds sort of colliding, right? You have the analytics guys who just want their answers out of the data. And then you have sort of the dev ops and the operations teams that have to deliver that. And as you said, as you move more to real time and you move more to sort of agile data, you really need, and the fact that the data sets are getting so big, and it's a polyglot world, right? Where you have to support multiple languages. To us, the logical answer to that is some type of a scale out infrastructure that's very easy to manage and very sort of dev ops friendly. And that's pretty much the description of OpusTac, right? It's a scale out, very oriented on a development ops sort of a perspective, very easy to manage, meant to scale out, meant to handle big data sets, meant to handle multiple workloads, accessing those different data sets, and be able to scale out as the demand goes and be able to sort of elastically do it, right? Just everybody is sort of elasticity is an assumption in the big data world. And so that's to us, that's a perfect storm. That's exactly what OpusTac is designed to do. So we think that and if you add into all of that, the fact that the vast majority of businesses, right, it's their data, they're trying to monetize it, they're trying to get value out of it, they're not going to want it outside their firewall, that's sort of a perfect storm in our mind for OpusTac. And you're thinking about OpusTac, what I like about OpusTac right now, remember the old OSI model of the seven layers of the stack? You know, the first two had to get done first, right? And then the TCPIP came up on top of that. So that's kind of really cool. So I see OpusTac really nailing that at the, I won't say bottom of the stack, but like that enablement, that infrastructure piece is really critical. Are you happy with where it's at in terms of the progress? I mean, I think generally people would like pretty happy with that right now. And then where's the work areas that you see the community really having to focus on? If you had to kind of say, okay, these are the areas to work on. Yeah, so I think you characterize it well, right? The things that are getting very solid is the foundation, right? So it's the compute, it's the storage, it's the networking. Those are all starting to get stable and those are our fundamental building blocks. The next sort of thing we got to do now is to make all the orchestration work, make the provisioning work, make the dynamic deployment work. That's non-trivial when you're talking about, but you can't do that until you get the basics done, as you said, right? So I think we're past the basics and now we're approaching that, now we're into the mix in terms of getting the deployment and the provisioning part all working nice and smooth. And then the next one is gonna be the workloads, right? Things like schedulers and all that is gonna have to come in and to basically take that workload, combine it with the provisioning of the infrastructure. And so that's where the heavy lifting is right now, that's where a lot of the attention is. So what do you think about the open data platform announcement? What's your personal take on, I mean, I've read how you're a purist, so let me take a personal perspective on that, that company perspective. Do you think it's gonna work? Do you think it has legs? Do you think it's full of smoke, hot air? What's your... Yeah, so it's very much the way you mentioned it before, right? We're the purists. So to us, it should be more about who shows up with the most code. And it's also about being open to everybody, right? In order to really do... So we agree 100% with the goal of the ODP. That's the same goal we have, right? Which is, hey, let's get a common base, let's all work on the same common base and let's go build on top of that. But in order to make that successful, to us it's the how that's different, right? We would like to see all the code in the open. No one group or one company controls that. It's truly open, everyone can contribute. And then you follow a process where it's an inclusion, you include everybody, you don't exclude anybody. And in order to be successful... That is a real critical piece. That inclusion not excluding is to me the number one criteria for success. Because that's truly in the open. I mean, that's our philosophy. We're all about open, not closed. We have some, we disagree with many people about their business models in the media business. You want to be closed and not include people. That's a choice. It's like religion, whatever you want to do. But if you're closed, a walled garden, not inclusive. So that's our philosophy. But again, if it's not done in the open, you gotta be open. So now the measurement with social media, you can't... People know what you had for breakfast before you leave your house. People, I mean, so it's all out there now. So like, compare and contrast what that does for the transparency, when you go back 15, 20 years, so IRC had back channels stop going on. But still, now out in the open, has that changed the game and open source in your mind, dynamics? Yeah, yeah. Can social media? Oh yeah, like you said, you know, it's more, everybody knows everything now. And so there's really no secrets. And then when you do try to have those closed door meetings, right? They're a lot of tweeted. That's right. They get tweeted before the meeting's over, right? And so the open data platform is a good example, right? That was leaking out there a couple weeks ago. And so you kind of can't do much behind the closed doors anymore. Just might as well just tell what your affiliation is. Right. And more importantly, to your point, you want to be inclusive. You want to include everybody. So what's the point of that? You're not really gonna, you're not gonna be able to control because everyone's gonna know what you're doing anyways. So why not invite everybody in, right? And if you're gonna invite everybody in, then there's already a business model for that. It's called open source. Where you put the code out in the open. No one company controls it. Everybody contributes. It's a meritocracy, you know, sort of bring your developers as the way you control it. Well, and now that open source has become so mainstream, I think there's an element that I'm watching is, okay, it might be, you know, the Democrats versus the Republicans or whatever we want, analogy you want to use. You know, what's best for the customer ultimately is the end game, right? So in the pure coding world, yeah, show up with code. Code will always trump cash. As Mike Olson would say, I love that line. Code trumps cash, meaning you can't buy leadership in open source. I agree with Mike Olson on that. But business model also is an issue, right? The business outcome will determine it. So, you know, maybe the diversity of models might be just an evolutionary game. So that's kind of why I'm watching it closely because it's certainly the bad tactics have all been seen before. We've, you know, talked to Cloudera Horton, works with you guys. We've all seen the dirty tactics. We know what that looks like. And so we'll be easy to spot those guys. So if it goes that way. Okay, final question for Red Hat. What's the strategy? Share with the folks out there. You guys are a leader in open source. You guys are the big grand poobah, if you will, for open source. You've got a ton of experiences as a company. A lot of chops. You guys have a hardened approach, 10-year SLAs plus on a lot of the enterprise stuff. It's super, super awesome. Took hard work to get there. What's next? What's the strategy of Red Hat? Share the strategy around the company, the plans, the code, the open source, quick summary and we'll wrap it up. Okay. Yeah, well so it's kind of more of the same in a sense. From a big data perspective, we're all about open. We think that's where all the innovation has happened, as you mentioned. We think that's the right business model for the big data industry. So you'll continue to see us build open interfaces, publish all of our APIs, be a player for every, you know, an entire ecosystem and provide sort of modular solutions so our customers can plug together the way they want to plug them together. They don't have to use all of our products. They can use some of them, et cetera. So you'll see more of that. Obviously agility is very important and also providing the infrastructure that's very dynamic, very elastic, like we talked about. We think that's open stack. So you'll continue to see a lot more hard work there from Red Hat and our partners and the ecosystem. And then finally, you know, as you've heard lots at the show, it's all about the application, the business value. The business value is all about drawing analytics out of the day, you know, having analytics that work against the data to get the business value. And so we're providing, you know, look for us to continue to work very hard on things like OpenShift and our Jboss Middleware to provide platform as a service so that customers can use whatever language they wanna use and develop in their tools that they want and yet do it dynamically and be able to deploy it very quickly. What about Red Hat's some of any visibility on what's coming on the agenda for that? Is it too early to kind of tease out a little bit of data on that? Yeah, no, it'll be the same as, are you, you know, what I just mentioned, right? It's gonna be all about open stack, it'll be all about set up. A lot of sessions, same deal. A lot of informational sessions, the queue will be there. Oh yeah. We'll be in Boston, hopefully the snow will melt by then. It was a report that said in New York there was the longest string of no murders in New York City in its history because of the cold weather. One of the comments was maybe the good bodies were frozen. But so the snow in Boston hopefully will be melted by then but we'll see. Greg, thanks for coming on theCUBE, really appreciate it. As always, a great conversation here on theCUBE. We'll be right back with our next guest after the short break. We are live in Silicon Valley, day three coverage, a big day to ask people to be right back.