 It's theCUBE, covering OpenStack Summit 2017. Brought to you by the OpenStack Foundation, Red Hat, and additional ecosystem support. Welcome back, I'm Stu Miniman, joined by my co-host John Troyer. Happy to welcome back to the program the two keynote MCs for the first two days. Jonathan Bryce, who's the executive director, and Mark Hollier, who's the COO of OpenStack Foundation, both of you. Thanks so much for joining us. Yeah, thanks for having us back. That's great to be on theCUBE. And thank you for the foundation. Without your guys' support, we couldn't do this. It's our fifth year doing the show. I remember the first year John Furrier went, they were like, hey, OpenStack's arrived, the cube's there. And now it's part of our regular rotation. I know our community loves it. Community, open source, big part of the show. I wish we had like two hours to kind of tease out all the pieces. But Mark, I got to start with you. You just did a live Q&A with Edward Snowden. Somebody joked, they said the quality and the sound was too good. He was sitting in the back room somewhere. Can you just tell us, how did this come about? And how do you make that work? Yeah, I mean, you know, pinch me as this real life. I keep asking myself, it seems kind of surreal. But, you know, just briefly, a lot of people have been asking, you know, how do we get connected with them? It's kind of a funny story, but basically, several years ago when the whole story came out about somebody from the government, from NSA had leaked these documents, but nobody knew who it was. And I was on vacation, it was in the summer. I forgot what year. I was on vacation with my family. And we were in the lobby of this hotel when we were on vacation. And I've been following the story with some interest and all of a sudden I see on the TV screen in the lobby of the hotel, breaking news, we're about to reveal the name of the leaker. And I look up and I'm watching it with intent. And it says, you know, here it is, it's Edward Snowden. So the first thing I did is I pull up my phone, I immediately look and see if EdwardSnowden.com is available. So I registered it, thinking, well, this might come in handy. This person just became the most famous person in the world, possibly. So it was available. So I'm furiously typing on my phone trying to register the domain. So I registered the domain EdwardSnowden.com. No idea what I would actually do with it, just thinking, if it's there, you know, this name is about to become really famous. So I registered it and it didn't do much with it. I just kind of put some Twitter feeds on there, just thought, well, you know, see what comes of this. And a little while later, as things developed, and he ended up, you know, in Russia, I was contacted by some of his team that said, you know, we're putting together a legal defense fund. It'd be great if we could host it at EdwardSnowden.com. You know, what could we buy the domain from you? And I was like, you can have it. You know, I'll donate it. I just sort of grabbed it because I figured this might come in handy someday. Just kind of as an impulse. And so they said, great, you know, thank you. Edward thanks you. Like, we're gonna really use this domain for his legal defense fund webpage and all that stuff. And then over time, I just occasionally would ping him and say, look, you know, the domain's free. You've got it. Like I want you to have it. It's not my name. I don't have any need. I don't have any right to this, you know? So you guys use it, but it would be great if you could come on the summit thing that we do. So this was like three or four years ago. And they were like, oh yeah, he would love to do it. You know, to thank you for kind of donating the domain. But each time we talked, it was always like the schedule didn't line up. And so I've been literally asking him for like six or seven summits. And this was the first time the schedule's lined up. And I didn't tell anybody because I thought, you know, this has never happened and this is a pipe dream and I don't want to promise anything. So it was only just a few weeks ago that we found out, okay, the schedule's lined up, it's on and got connected from there. And you know, he's obviously an open source person has a lot of passion behind that. We thought this is pretty interesting for our audience. So it worked out. All right. So Jonathan, let's reset for a second here. And step back. One of the things we'd love to see is the foundation is self-aware. There's always that balance when you get into is like, you know, you don't want to read the press or things like that because they don't understand what we're doing or where we're going or things like that. But in your opening keynote and throughout the show, we called it, it's a little bit of a reset. If you think about, you know, where people thought OpenStack was and where it was going three years ago, it was like, oh, the Amazon this or the cheaper VMware or how that is, where it is, where it's going, who's leading, who's involved, you know, winning and losing type stuff. You guys did a good job of laying that out. So congrats on that. Why don't you take us in a little bit in what messages you guys want to get out this week? Yeah, I mean, I think that you're right. We are very self-aware. And I think that some of that comes from our role is, you know, at the foundation we are not selling a product. You know, we don't have anything to sell off the back of the truck, so to speak. So what we actually really do care about is just kind of moving the state of the community and the technology we produce forward. The thing that's great about that is, you know, we can look at the portfolio of technologies that we have. We can look at the things that are in the market. And if we see a shift there, it's not like we have a $500 million line of business that, uh-oh, you know, we need to keep milk in this cash cow and not, and sort of like turn a blind eye to these changes over here. And so I think, you know, over the last couple of years, I talked about a shift in kind of what private clouds can do now and how they're built and operated. And we've seen that, and we've sort of been teasing that out a little bit at previous summits, whether it's demos with Kubernetes or different integrations with Cloud Foundry and other things like that. But what we decided this time is, coming out of last year, there's a lot of news and what we saw really kind of picking up is there would be these rumors or misperceptions that somebody would put out there, you know, and not based on facts, not based on reality. And we were like, you know what? We can't just sort of try to subtly hint at what's going on. Let's just go out there and actually address the state of things. And I think what you mentioned is actually what's at the root of a lot of these misconceptions is people look at open source now because so much technology gets developed that way. They look at it and they expect it to be like the old world of IT where you need to have Microsoft versus Linux and you need to have Oracle versus MySQL. And actually what we see is just the cloud overall is growing so quickly. Public cloud, everybody believes that's growing. What we see is private clouds are growing and we see that servers, there are more servers this year than there were last year. There are more virtual machines this year than there were last year. Far more containers this year than last year. All of these technologies are growing. And so it's not a zero-sum game where, you know, in order for open stack to succeed, AWS has to lose. And, you know, I mean, that's, I think that we feel that way and we see that but we realize that this is, we need to just go at it directly. Mark, I've heard good feedback from people when core, where it is, how it's matured, people like the component piece. So you'll be able to take some of the digital pieces, which, my understanding, they could do that before. It's just kind of being highlighted a bit. We talked about some of the open source days and, you know, cloud foundry, Kubernetes. The piece where we've heard some people poking holes in is what BigTent we discussed last year. You know, BigTent, have we poked a hole? Is it dead? How do we reposition that? Yeah, I mean, it's a great question. I think first of all, one of the things that sort of just this strange stroke of luck that maybe turned out to be bad luck was, you know, one of the few times when a handful of developers kind of went off and organized something, gave it a random name and the name really stuck. You know, it actually was almost too good of a name. So you heard BigTent, everyone's just rolling off the tongue all the time, BigTent, BigTent, Mente. So everyone had to have an opinion about it and was like, this is a huge change and it really wasn't meant to be a huge change. It wasn't even meant to be like, sort of broadcast that widely to everyone who's like just observing OpenStack. That's just kind of what happens, you know, people talk about it. And so I do think that we are entering a point now when we're thinking about composable open infrastructure. Yes, you need to have different components. You need to be able to pick them. But we're also getting more serious about what things need to exist in OpenStack. I talked about that a little bit this morning. Not every single thing that we've ever launched needs to continue to be an OpenStack project. And so, you know, whether you call it the BigTent or not or if you give it different names, you know, the reality is, you know, we need to adopt and integrate technologies from other communities. And that's a, any open source community out there is potentially developing something really powerful. Did you mention the XED thing this morning and you're, I can't remember if you... Yeah, I mentioned it briefly, but you know, a perfect example of this is a lot of OpenStack services have said, you know, we need a distributed lock management function like in order to evolve as a service. Where should we go build it? How are we going to write it? And then this kind of culture of, well, hold on. There's, there are a lot of them out there. They're proven. What about XED? And so the forum, which is the first time we've really had a dedicated space at the summit for both developers and operators to be in the same room, not just like next door to each other. They had a discussion yesterday on this and they said, yes, we're going to go forward with XED. That's an open source project, very proven. It solves this particular function. It's not developed inside of OpenStack, but that's who cares, you know, it's open source. We can work, we can be friends with anybody who builds great open source software and let's not reinvent the wheel. So I think that does represent a bit of a shift in the philosophy and culture and OpenStack of not trying to like, you know, just build every single thing from scratch because that's not the best thing for our users or the market. I think the ecosystem message and the landscape message came through really clearly. This is my first OpenStack summit. I was very curious about what is the shape of OpenStack? Where does it fit in? And, you know, talking about the upper layers and Kubernetes and the app layers and now talking about the overall landscape. Right, you know, why are we right? That's something like XED, right? The whole ecosystem has grown up around OpenStack during the seven years you guys have been, the whole foundation has been working on it, all the members. One thing that impressed me, we are kind of post hype cycle. There are real customers here. There are people building their first clouds right now on OpenStack. Could you talk a little bit about just the community in general, the composition of it and the actual real use cases that we're seeing happen? Yeah, so we had some new companies that spoke here for the first time. GE was one. The US Army Cyber School was another one. We had some companies that came back as well. But I think that you hit on a key point, which is the maturity of the software. You know, a company like GE, especially in their healthcare division, this is a highly regulated company. It's probably the most regulated company out there when you consider the things they do with aviation and nuclear power and healthcare and finance and all of these things. They don't take those decisions lightly at all. And so I think that is an indicator of that maturity. And what we see in the makeup of the community is a broader set of industries than ever before. We had strong representation among IT companies early on and continued with that. But now we have industrial companies, we have manufacturing companies like Volkswagen, BMW, a number of car manufacturers and defense companies. And I think that that kind of plays into that. I think the other thing that we've seen, when we talk about the open-stack community and the platform overall, we think of it as an ecosystem that sort of has three main parts. There's the users, which that's why we exist. We create software for it to be used. There are the developers who are doing that. And then there's the ecosystem of companies who create commercial products and services. I think that that's actually just as important right now with the phase that we're at is how that is also reaching maturity. In the early days of open-stack, I think that we had a lot of startups and we had a lot of activity, but the market didn't know how to consume it. It didn't understand what it was. And I think that actually kind of scared off some companies and it made it a little bit more confusing. But as you get a few years into that, some of those companies succeed, some of them don't succeed, but what you arrive at is a clear understanding of what the market wants, how the products should shape up. You get companies that stop trying to build it all themselves kind of along with the not invented here and they partner with people who know how to do open source or they come up with new delivery models. So I think that actually just as important is the maturing that we've seen in the commercial ecosystem because that leads to sustainable business models for these companies like Red Hat and Rackspace and others that then drive the development, but it also leads to clear adoption choices for users. One of the things that I came out of last year at the Austin Summit was just where OpenStack fits in in this hybrid world. I think about GE, Rackspace, Red Hat, all of those companies clearly span both sides of it. Back to that like winning and losing discussion we had at the beginning, it was like always public cloud versus the private and the infrastructure piece and we know it's a multi-cloud hybrid cloud world. How do you see that fitting and in the conversations? And the other piece on that, I see a large number, it was 74% of deployments according to your latest survey are not US which kind of is the inverse of we see such North America's where we have a lot of public cloud adoption. So does that fit in, what dynamics maybe start with you, Mark? Yeah, I mean a couple of things I would say that what we're finding is a few years ago was like are we going to do cloud? Okay, now it's yes. Then it was which apps is going to be, it's going to be as many as we can get. And now it's, are we going to do public or private while we picked one? Now it's okay, yes to everything. It's going to be cloud, we're going to put as many apps as we can and we're going to do public and private. So what happens next? Now it's a question of where, where do you place each workload? And some of them belong in the public cloud, some of them don't. Economics plays a big factor, performance, compliance, all the things that he said, the three C's, capabilities. And so I think that that's the next discussion point that's happening inside of these board rooms and CTO's and IT leaders at the major companies is how do we get a sophisticated strategy for where to place the workload? And then in terms of the geographic dynamic, I think one of the things Jonathan hit on yesterday is that we start, it's just the nature of open source that you never know where it's going to go. You just have no clue. Really any new technology development, the market's going to go somewhere you could have never predicted. Like the crystal ball is kind of dead. It's really road maps are almost kind of obsolete. It's like you need to create a structure for how you respond and adapt to change because you know it's coming. And so what's happened with OpenStack is it's been used in all these new and different ways and part of that's geographic. It's used to power cell phone networks and all these different countries. It's being used to fit within regulatory requirements in certain countries and data locality, both for performance and other reasons. So I think that's why you see it. It's a big world out there and more than 74% of the world doesn't live in the United States. So I think we're closer to the real percentage out there. I want to jump in with one thing that you said that I might disagree with slightly. Okay, let's have a debate on the right. Yeah, well you said that these are the conversations that CTOs and CIOs are having. It's a strategy about how to do it. I think it's a conversation they should be having. Okay, fair point. But I think that what we see is we see a lot of companies. After they hear this. Yes, after they hear this. Start talking about the right thing. But I think that we see that, but we are kind of on the front edge of cloud adoption in the OpenStack community. I can see your point, sir. And I think that one of the issues that we see still is that people are thinking about it too simplistically, almost, and as Larry Ellison famously said, the IT industry is the most fashion-driven industry out there and I think that right now there are a lot of companies that they still think that there's some shiny object that's going to fix it all for them. And right now it might be public cloud or containers. They've heard this word and they think that's never happened. Never happened in the history of IT ever before. There has never been a technology that came along and fixed the stuff before it. They all get out of it. And I think so yes, you know what we talked to, we were talking with a CEO just this week and it was really interesting to hear his perspective because he said that he actually thinks that the pendulum is going to shift back towards private cloud for people who run any significant amount of software. And he goes, I know that is not a popular viewpoint right now and if I said that to most other technologies, C-level execs, they would probably disagree with me and go no, cloud-first containers. But I think that just the fundamentals behind it over the next few years, I think, I don't know if it'll shift all the way back, it may, who knows. But it's definitely something that I think is going to change from where the current fad might be. All right, we'll have to have you back later to talk about how public is now moving to Edge and Edge of course lives as the new, Edge is the new data center is what they have. I do have one final question for you before we let you go. That whole new shiny stuff for the last couple of years, I've been hearing everybody's like, well, containers are going to subsume and take over and DockerCon will be the new thing. Oh wait, Kubernetes is just going to dominate and take it over and we have KubeCon and the CNCF and there's lots of Linux foundation shows that do partnerships with what you're doing, Cloud Foundry Summit and all these other pieces. So what do you see as the future for the OpenStack Summit? Does it get pulled, it's being pulled into pieces but for the show itself or the foundation and how it fits with that whole broad ecosystem of open source? Well the OpenStack Summit has always had some specific purposes and again this gets back to the fact that we are an open source community and a foundation built to support that open source community. So the primary purposes of the OpenStack Summit are basically to strengthen those three pillars that I talked about earlier, especially on the software angle. Mark mentioned that this time around we are doing what we call the forum. We used to have the design summit here and we actually split that into two parts. One that is, that's very technical and it's really gets down into implementation details. That's split out into a separate event. It happened in February, it's going to happen in September. And what we did here is we set up time where developers and operators can get together and talk about strategic issues. So instead of talking about how do we fix this issue on line X of file Y, they're talking about what should we use for distributed storage and lock management? Should we do at CD, should we do like, they're having more strategic conversations. So that is a very critical piece for our community and for the people who run on it. We do a lot of education here. I mean, I think that what we've seen is that the OpenStack Summit is becoming more focused around users and kind of the strategic needs of them as we build out the technology versus what it used to be, which it originally started as a hacking event for 75 software developers. So that's where I think it's going. And just to address the other point, all of the other open source projects, a lot of them are here and we go to their events because again, like we've been saying, it's not a zero sum game. What we care about is that there are open alternatives and that they work well together. And one of the things that I think we've seen and we've seen it proven over and over again with OpenStack is that getting communities together in person, those high bandwidth interactions are actually really critical to getting work done and making things happen. So I think they're all valuable and we're going to continue to participate in all of them. Jonathan Bryce, Mark Collier, really appreciate you joining us. I'm sure we'll see you at many of those other shows that theCUBE will be covering throughout the years. Stay tuned with us. We've got lots more coverage here at OpenStack Summit 2017 in Boston. Thanks for watching theCUBE.