 from Seattle, Washington. It's theCUBE on the ground, covering KubeCon 2016, brought to you by the Linux Foundation and Red Hat. Here's your host, John Furrier. Hello everyone, welcome to theCUBE special on the ground coverage of KubeCon or CloudNativeCon. This is an event Seattle booming with attendance, great growth from last year, and we are here in Seattle covering it all. My next guest is Dan Kahn, who's the executive director of the CNCF, which stands for the CloudNative Computing Foundation. It's a mouthful, but it's super important. Part of the Linux Foundation, welcome. Thanks so much, really glad to be here. Yeah, so big fan of what's happening here. One, the event's awesome, great uptake from last attendance from last year. Yeah, unfortunately, maybe a little too much. We're a little crowded in the foyer and a little bumping on the way into getting the restroom and everything, but it's one of the challenges of fast growing technology space is trying to figure out a year ahead of time, what size space to get. And how many people will squeeze in without getting the fire marshal on your back? Exactly. Certainly this is gonna be a great one because the hallway conversation has been spectacular and not only the excitement's pretty strong at tech events like this because they're developers, there's a lot of collaboration going on, but you have a kind of an air of really forward thinking entrepreneurial kind of thinking going on here. And I haven't seen that in a while and I think that's one of the main things that we're seeing that came out of the containers, Kubernetes, I would say the unveiling and the clarity of at least a path. Yes, absolutely. And the importance of that. So that's been super important for the Delbar community. Now making that a part of the foundation and open source has challenges. So that's what you're doing. So give us the plan, what's the strategy? Sure, so I'm actually relatively new to the space. I just became the executive director five months ago and this is somewhat of a coming out party. This is the first big event that we've run as the first cloud native con and it's really just been extraordinary. I mean, I'm thrilled to see the range where we're getting some of the biggest companies in the world of the Cisco's and Wallways and IBM's and Red Hats and such and then tons of startups and a lot of real diversity in the end users as well. Of startups looking at Kubernetes, massive companies just saw a great presentation from Ticketmaster about them having 50 year old technology that they're moving forward and putting into containers. So and the growth of the market, one of the challenges is to kind of, not so much be a chess player, but be a gardener if you will, kind of let the flowers bloom if you will. And that's the challenge because open source is very opinionated, but there's also a lot of passion involved. So how do you look at what's your philosophy on establishing kind of a rules of engagement? How do you foster the innovation? Certainly the market drivers are for more growth but people have inhibitors on the enterprise we hear about support and these things of that nature. So how do you enable that? What's your strategy and what's your view? Sure, so CNCF is a very new organization and my goal on it is to look at a lot of the giants that have come before us have liked the internet engineering task force and the Apache Software Foundation and OpenStack. And my goal is to try and learn from them and ideally to try and make entirely new and different mistakes as opposed to the ones that they may have made in the past. So one of the things that's a little unusual in our setup is that we very much separate all of the technology decisions from the business decisions. We have a governing board of a bunch of the biggest technology companies in the world, the ones I mentioned plus Google and Samsung JustJoin which we're very excited about, a number of others but they can't actually adopt projects in. So we have a separate group called the Technical Oversight Committee which is some of the top architects in the cloud space. So we have folks like Ben Hindman of Mesosphere and Salman Hikes of Docker, Brian Grant of Google and six others and that's the group that looks at new projects and evaluates them and talks to them and decides whether to adopt them into CNCF or not. And we feel that that separation is really critical so that the technology decisions are not being biased by the business one. Yeah, it's always hard to foster growth and innovation around business models conflicting with the technology enablement. That's really key. Great to see that decoupling. So on the business model side, what are the thoughts on things that you've learned and observed learnings that you've had in your past career and applying that now? I mean, the rate is on open core to a patching, GPL. We saw some things going on there. So there's like all kinds of different approaches. Are there any thoughts of the winds blowing any way it's way or the other? Sure, I was previously the Chief Operating Officer at the Linux Foundation between 06 and 10 and I definitely think you can, CNCF is part of the Linux Foundation. We took that model of saying the technology decisions need to be separate from business ones. One thing that's interesting to me is that when I was last in this space 10 years ago, people were perfectly fine, you know, the Linux Tunnels GPL, people were fine with free licenses like MIT and BSD. Since then, and for this group, there is an enormous focus on the Apache license. And the reason why is the fear of submarine patents. And so the whole goal of CNCF is for us to be an intellectual property, no fly zone, that you can have all these companies that compete very hard in the marketplace, but they can come together and collaborate and share their ideas and their technology without the belief that a couple years later, someone's gonna be able to trick someone else in with the lawsuit and win that. And the Apache license is really the industry consensus right now for best practices. It's interesting because that no fly zone gives the freedom for the creation and the invention side of it. The patent thing is always worrisome, but in general, there's also the business model down the road kind of approach, which is let's go innovate. Apache's done great on packaging, and if someone gets some traction, it fosters the community aspect as well as startup, maybe thinking about packaging. Now we have an advantage that we're not, unlike OpenStack is an example, we're not trying to come up with the projects ourselves. What we're actually doing is scouring the cloud native landscape, talking to different groups and saying what do we think is the best in-class project out there. And in some cases it's more than one, but today we just announced the fourth project that's added to the CNCF. So we have Kubernetes, we have Prometheus, which is a monitoring application, OpenTracing is a tracing, and then today we just added FluentD, which is a logging solution. And this is the idea that if you have dozens or hundreds of different applications and projects that are each producing a log stream, and then you have perhaps dozens of other applications that are consuming it, you don't want to have an M times N problem of creating adapters for all of them. Instead, you can plug them all into FluentD, it has over 300 adapters for different solutions out there, and that provides one comprehensive approach. But what's interesting is that we don't need to win over the community and say, oh, here's this project you may not have heard of, there's actually over 2,000 users of that today, but by having them here at CNCF, showing how it plugs into other technologies of ours, we think we can help them get. And you're cross-pollinating. Exactly. You're letting it bubble up, and you're not being a dictator. That's exactly the metaphor. A little dictator. But okay, and back to the project side, this is awesome. So you have some gravity around these projects. Is there any cadence or expectation, or is it free for all in terms of the velocity of adoption of projects that the technical committee will oversight? We would love to be at the pace of one a month, and I don't know that we'll quite get that fast, one big change that we're hoping to make in the next month or two. When our first two projects were Kubernetes and Prometheus, those are two of the fastest growing, best respected projects on GitHub right now, we didn't want to have such a high milestone for every other project we considered. So we're adopting what we think we're gonna call an inception stage of earlier projects that we're gonna sort of try out, but they have to essentially prove themselves within 12 months. And hopefully that'll allow us to keep a pretty good velocity where we think there's a fantastic number of projects we think is a community we can support. Yeah, let people fight it and let it surface stuff and let people kick the tires, right? Exactly. Incubation period, basically. Right. What about the forking and all the battle cage matches that go on? How do you want to handle that? You just let nature take its course. Is that philosophy there? Thankfully, when we look at the space, and this is really coming out of the Linux space as well, anyone can fork, and of course it has a slightly different connotation now with GitHub where when you make a change, you fork it. But there's also just a massive centrifugal force pushing people together. And when you have a really high velocity of changes, the idea of forking and you would lose out on that becomes a lot less appealing. And so, so far, thankfully, all of our members and everyone in the community has really been on board on having a single head on working together to have that consolidation. We just had Richard Kaufman on from, I think Robert Kaufman, I mean from Samsung, he was talking about that the number two contributor is other. Yes. And it's a nice balance to the whole critical mass feature. Yeah, it's an incredible accomplishment because for Google to pull in enough people that they're no longer the majority contributor is something that we're thrilled with. Yes, great to see you have Richard Kaufman. Richard Kaufman. Google is the number one contributor. Are you worried about that? Maybe they've been certainly good actors in the community. I mean, they had MapReduce and like Cloud Air run with it, look what happened with that. So, you know, we kind of all know the backstory of Kubernetes. They're kind of letting it bloom on its own. That's consistent with their current posturing? Well, I don't think they want to have another Cloudera. Yeah, that's what, which is why they embraced Kubernetes. But I definitely don't think it's fair to say that they're doing it on their own. They're still the largest contributor of any one company. And they have a massive amount of resources and I think they see it as a really key technology. It's something they made in the early school. What I was referring to is that Cloudera kind of took MapReduce under their wing and made a commercial venture out there. Oh yeah, absolutely. Google didn't want that. No, and they, I mean, the way I think about it is they had this technology a few years ago. This is definitely oversimplified. They could have kept it as a proprietary in-the-house thing like Amazon Elastic Container Service. They could have made it an internal open source project like Go, or they could have just created a Kubernetes foundation that allowed other people in, but they still controlled it. But instead, they were really interested in working with the Linux Foundation in creating this cloud native computing foundation that was always designed to be much more than just Kubernetes. And that really was about trying to push the project out of the nest. But I will say that my understanding is they're still, see that as an absolute core for their business. Yeah, I gotta give Google props out there for that because they did do the right thing there. They put it out in the open. They did align and they could have landcrabbed that in a different way. I mean, certainly not the way that one was. But a final question on this event, KubeCon or KubernetesCon, it's KubeCon, however we call it. Not to use it, the Kube. This Kube product, it's seven years and might be trademark infringement, but look at that later. There's a K there. There's a K, it's still causing a lot of confusion. But that aside, cloud native con also is in conjunction. This is part of the expansion you were mentioning. Talk about the vision for the events you got one in Berlin bringing up. And certainly you could have had, probably at least a few more thousand people here for sure. Well, certainly a few more hundred and we do feel a little bad that we didn't quite aim high enough. So our vision going forward is that we have cloud native con that represents all of our projects and that KubeCon represents the biggest part of cloud native con. So it's multiple tracks. It's what a ton of folks go for, but we think that it also gives us a chance to expose those people to our other projects. And by the time we get to Berlin, we're certainly hoping that we have another two or three or more projects that are added in. It's March 29th and 30th. And then we also announced that we're gonna be in Austin in early December. And I'll say that for both of those events, we're tripling the capacity from what we had last year. So we're hoping not to be so crowded. I was talking to Andy Jassy last night. We had a one-on-one with him and he was talking about the first re-inventing. Didn't think he can get, you know, 4,000 people there as backed. I think you might have to look at more capacity potentially at this pace. It's the hard question is we'd actually like to be signing contracts for 2018 and it's just really hard to predict what the right size is to get for that. Because I feel terrible about the fact that we did turn people away, especially end users that we'd like to be introducing to this space. Yeah, well, I can attest people watching this. Definitely a fire marshal issue because it's really packed. That's why we're in a separate room here. The sunlight in the background earlier. Normally we're on the show floor with the queue, but yeah, every space is taken. Always a jamming. The other thing I'll mention though is that we are also interested in going out and reaching customers and vendors where they are. So we're going to have a booth at AWS re-invent and we're looking at other conferences that we can be at to help spread the cloud native word. Well, we're certainly going to be over 100 events this year. So let us know where you're at. We certainly bring you guys on. I'll give you the final word. Tell the folks why Kubernetes is so important. Why is this movement? Why are people so excited here for the folks that couldn't make it? Why is it, what's the vibe? Why is it important? And what's the impact the customers in the industry? So the belief is that if you're deploying a new modern software application that putting it into containers, using an orchestration platform like Kubernetes, dividing your app up into microservices is really such a benefit in terms of optimizing your resources and tying into a whole rapid development process, continuous integration, continuous deployment, that not doing it almost makes it impossible to compete. And so we think there's just a ton of momentum around containerization and orchestration. And the speed of the innovation is one of those things. If you're not on it, you fall further behind. Exactly. And it takes more energy to catch up if you try to do it by yourself. That's the benefit of the collective communities and any highlights open source. Right. Big time, in terms of successes. Thanks so much for coming on and sharing the perspective. Congratulations and sorry for the folks who couldn't make it, hopefully this video will help. This is theCUBE here in Seattle for special coverage of cloud native con and kubecon here in Seattle. Thanks for watching. I'm John Furrier. Thank you. Thank you.