 Here inside the queue for exciting wrap up of the Enterprise activity, OpenStack Enterprise Forum. O.D. forms the hashtag. Go to crowdchat.net slash O.D. and pitch me. I'm John Furrier, the founder of SiliconANG. I'm John Michalho. Dave Alonso, co-founder of guvon.org. Our next guest, saving the best for last, Chris Kemp, founder of Nebula and OpenStack, former CEO of NASA, making a dent in the universe by democratizing cloud computing. That's his Twitter bio right now, as of today. Chris, welcome inside the house, inside the queue. Appreciate you coming back. Two days in a row, we had Cole on yesterday. I think he's great to be here. We had all the luminaries. It's funny, Dave, you've been here over so long, you meet everybody when we knew them when. So now that they're all famous, Chris and Cole, big journey for you. I mean, come on. Chris, how do you feel right now? Enterprise is hot, packed house, Twitter's on fire. Yeah, it's great. Crowdchat's going crazy. It's great to see hundreds of people show up for an event that's just focused on Enterprise Cloud and just a couple of companies here. I can remember a couple years ago, we had a room full of 15 or 20 people at the first OpenStack meeting and now we have thousands of people that gather around the world to participate in the community. So it's just incredible to see how fast it's grown. What is the deal with Enterprise Cloud? We were just at the Open Compute Summit earlier yesterday, covering it this morning. We didn't have the cube there today. But you have this homebrew computer club mentality right now, inflection point, massive sea chains, transformation, growth market, bubble, whatever you want to call it. It's massive investment, great action. Yeah. But what's the big aha? Why Enterprise Cloud? Why is it so freaking hot right now? Well, I think Enterprise is realizing that they have some big challenges in IT. And if it's an insurance company that's trying to figure out how they can collect all that data from cars to figure out how to price premiums or to better understand their customers, it's advertising companies that are trying to understand how to take all the click streams from the Internet. Every place you look, financial services companies trying to do a better job of understanding markets, you have to build massive scale-out computing infrastructure. And I think Tim O'Reilly actually said it best when he did our company launch video. You know, just a few years ago, no one would have imagined that a company would need a Google scale compute infrastructure. Now everybody's building them. So you've got projects like Open Compute that are democratizing the hardware and OpenStack democratizing the software stack. And obviously there'll be companies like Nebula that help deliver some of this technology to the Enterprise. I want to ask you a question because this is my theme for the week because I was really, I'm an older, one of the older guys, Dave and I are a little older. I'm in the generation of the Mac guys, about ten years younger than the guys who built the original Mac. And that was, you know, the computer science. That's when we cut our teeth into coding, that kind of generation. It made me think about those guys change the world because they built stuff for themselves. If you look at the stuff going on now, what you're working on, people in this room, these are guys that are building it for themselves because they know that that resource that was elusive before, that was controlled by the big guys, is now fully available, tax-free. So I want to get your perspective on that. The modern-day version of what that homebrew club did for the computer revolution is kind of going on, I would say cloud, whatever market you want to call it, I just globalize it and say cloud computing being represented of all the nuances of convergence in mobile. What is that, what's that world like? You're in the center of it. Yeah, I think we're at the beginning of a new generation in computers. I think there was this mentality that computers were these big, large-scale systems that were primarily focused on large business problems. The personal computer disrupted that and you had companies like Apple and Microsoft build these very small systems which ended up being incredibly successful and that transformed enterprise computing. So if I'm building a large enterprise-scale computer, I'm going to start using the components that exist in the PC. What's happening now is, it's happening all over again. We have mobile phones and ARM processors and DRAM chips and flash chips that are coming and spilling over from these consumer devices into enterprise platforms and it allows us to rethink the way we do software. Let's not just run software on one big computer, let's run it on thousands or tens of thousands or hundreds of thousands of small computers and I think we're at the beginning of this revolution and it'll be a 25-year journey. 25 years from now we'll still have software running. I'll be dead, you'll be like... Yeah, but I think it'll be the new mainframe. All the software that's running today will be running in a rack of VMware somewhere 25 years from now. So it'll be exciting to see. So we're at the beginning. In your view, we're just tinkering around. Early days. We're wiring it together with infrastructure. Yeah, early days. So when we first met, you told me, and I wonder if you could just do the bumper sticker version of that story. You told me a great story about really the roots of OpenStack inside of NASA, how you're able to successfully extract that from NASA, which is great. Maybe it trashes the US government. So many great innovations come out of the US government. It's unbelievable. So the short version of that story, and what led you to Nebula? Did you envision that you'd actually be doing Nebula? I got this picture of a box. I'm like, wow, it's Chris's company. Yeah. So when I was at NASA, we had a problem to solve. We wanted to share NASA's vast treasure of knowledge with the American public. And if we looked at how we could do that, there's not enough funding that we could have ever gotten to do that. So by partnering with Google, I had a view into how you could manage very large-scale computing infrastructures. We partnered with Microsoft, and we had another view on how you manage infrastructure at the time. And we ended up using these partnerships to build a very, what we'll call a prototype of a very inexpensive, scale-out computing environment. We did it in shipping containers, and the impact that that had was, you don't want to have people managing servers in shipping containers because people don't like to work in shipping containers, right? And so it had the added effect of automating all the servers in that shipping container externally. And OpenStack really emerged out of that work. And we've obviously seen it go to incredible places since then. OK. And then you made the comment. I tweeted out, Chris's tongue-in-cheek comment was, mark my word, someday that Amazon will be integrating with the OpenStack APIs. But we joke about it, but what's your position on that? We all know Randy Byers is very vocal on that. We heard some pretty couched opinions today on the panel. Where do you stand on that? I won't be couched about it. I think that having OpenStack APIs allows this community to innovate. It allows OpenStack to be something different than Amazon. I also think that what Amazon is doing as a company with Amazon Web Services is they're listening to customers and they're innovating. And I think it is a valid way to look at a segment of the market. I think it's looking at a segment of the market which is mostly small companies building new applications. And OpenStack is attempting to bring a lot of existing technology into the enterprise. And so I do think that it's fair to say that there's a different set of opportunities that exist when you bring a cloud into an enterprise. And I don't think Amazon will see those things. And so I think having the OpenStack APIs be discreet and can innovate on an independent trajectory will create more innovation for private clouds and for competition, frankly. So I do think both are important and that's not hedging. But I also think the Google Compute Engine APIs could be important as we start to see people do incredible things on that platform. And the market will decide, do you feel like Amazon's ambitions for the enterprise are overly ambitious and Amazon's being somewhat naive about the enterprise? I mean, Amazon, we've talked to Andy Jassy and he's flat out. I'm going after that trillion dollar opportunity that is the enterprise. There is no workload, no situation that we will not pursue. I'm very confident of that and putting a lot of resources behind it. Do you not buy that? You know, I think that when we talk about making large enterprises, giving them the power of a cloud platform, it isn't just about technology. It isn't just about making sure you can plumb the network. Those are important things to do to make it possible. But what you're dealing with is you're dealing with a cultural transformation. You're dealing with a different way of thinking about building software. And I think most importantly, you're dealing with a lot of existing applications that are just not going to run very well in the kind of ideal cloud architecture that we see the Amazon style cloud companies leveraging. And that's where the economic advantages of cloud are. If you're not able to elastically scale up and down your infrastructure, you're not able to take advantage of the economic benefits of cloud. And frankly, a lot of applications, I commented earlier, what I think is brilliant about VMware strategy, and they're a very smart company and they have an incredible product, they're virtualizing everything. They're virtualizing the network, they're virtualizing storage, they're virtualizing compute. And so they're creating an effect, and I'm a Star Trek fan here, but they're creating a holodeck for software. And you put all your software in there, and it'll run like it's 1995 in the year 2025. And you can change the hardware out a couple generations of hardware underneath it. And that's a good thing, just like mainframes are still running. This is our fifth year of theCUBE. This is the beginning of our fifth, I guess, season, I'm going to call it. But in 2010 was our first kind of initiation of theCUBE. We were at VMworld, one of our first early shows. Maritz launched the stack, right? And that was the software mainframe we all remember those days. A lot's changed since then. I mean, the vision, I mean, I was drinking the Kool-Aid, passed the peace pipe, I'll smoke that all day long. You know, beautiful. However, the top of the stack just never evolved fast enough. So you saw some end-user computing, you know, kind of misfire multiple times. Now it's kind of coming home with mobiles, okay, matured. And then Maritz takes all that out of the pivot, makes total sense. On PowerPoint, makes a lot of sense. So I want to ask you, because you're also smart, you were also, you know, NASA, and then with Sprite and Founding, Nebula, knowing what you know now, given some of the market conditions of what's advanced faster than others, if you can go back and do a mulligan with Nebula, what would you tweak a little bit? Because some things just didn't materialize fast enough, some were held by other factors. What would you do differently? I mean, obviously if you could go back in time and you could look at everything that's happened over the past few years, one thing I'll say is I'm very optimistic about the future. I'm optimistic about, I think there are visionary CIOs and CTOs at America's most successful companies that understand that the only way that they're going to be able to win is to power their most critical applications on extremely inexpensive-scaled infrastructure. And this makes so much sense to me, right, coming from NASA. But I find that there are a lot of companies out there that have CIOs that work for CFOs that have big challenges to deal with, with their existing applications. And they're also optimistic about where they are in their journey of virtualizing all these applications. And I think many of them say, well, we've virtualized 80% or 100% of our applications. And I don't think that's right. I think they've virtualized maybe 50% or 60% of their applications. And I think that's a good thing for VMware's market share. It'll continue to grow. But I think in the long term, when we see the incredible advantages that big financial companies or insurance companies or media and entertainment companies get when they can use something like OpenStack on OpenCompute, and Nebula's going to be a part of that, I think that you're going to have winners emerge. And it won't even be a fight. So would you do anything different going, knowing what you know now? I'd make it easier. And I don't know if there's any way we could have done this differently because we have, frankly, what's been called an HP strategy. We built a system that could scale kind of big and kind of small, but not super small and not super big. And so then we can start there and then scale bigger. So I think if I had to do it over again, seeing what some companies have done with rebranding a small server and selling that as a data center in a box, maybe making Nebula even less expensive and even easier and even simpler. We're so easy and so simple, but to just never underestimate the power of simplicity. So we're getting the hook there from the folks here, but I want to ask you, okay, now let's look forward because you're an optimist and we just want to get that question out there because the market's changed. We had predicted some things would move faster if they didn't. Going forward now, you get the chess board. You're looking at the chess board of where we are now over the next 12, 24 months. There's going to be some sprints and some straight and narrow, maybe some curves. You've got to slow down a little bit. Where do you see the current landscape now with Nebula, Visa, V, the marketplace? Well, I think the key is to allow customers to more easily run applications. So what will happen here is this ecosystem will emerge. OpenStack already has a huge ecosystem around the infrastructure. What now needs to emerge is the ecosystem around running things on infrastructure services. And so you're going to start to see every software company that's ever existed solving every problem you can imagine, re-architect on cloud architectures. And these applications will run on Nebula private clouds, they'll run on public clouds, and they'll run in both. And I think the really exciting thing for me is just watching as a whole generation of software evolves to be completely dynamic and elastic. And that, I think, will ultimately make this transition. That's where the knee and the curve will be when we start making it possible. When you think about your first Mac or your first PC, what would you do with it if no software ran on it? If everything you ran, you had to write in hypercard or basic. Remember when Microsoft had the basic and Apple had hypercard and that little... So this is where we are with OpenStack. What we have is we have heat. Heat is the new hypercard. And we've got Chef and Puppet, and everybody has to do everything. Things will change when all the software that everybody runs in the enterprise runs on cloud. You're a Star Trek fan, I'll end with this quote from the founder, Screnta, he's with the search engine, I'm blanking on the name, but he said, everything that was on Star Trek will be invented. Maybe the holodeck might be a little bit different. The warp drive is going to be tough as well. Warp drive will be hard. Beaming for warping. Big Star Trek fan. Chris, thanks for coming on theCUBE. Really appreciate your insight. And again, your perspective is very relevant. You've been there at the beginning. You've seen the trend. You've seen the chessboard. Great market. Congratulations on your success. Appreciate you coming on theCUBE. This is theCUBE, our special exclusive coverage of OpenStack for the enterprise. Breaking out. O-E, forum, that's the hashtag. Follow us on Twitter and crowdchat.net. We'll be back with a wrap-up after this short break.