 Okay, welcome back everyone. This is Silicon Angles theCUBE, our flagship program. We go out to the events and extract the signal from the noise. I'm John Furrier, the founder of Silicon Angles. I'm John Mike Co-host, Stu Miniman filling in for Dave Vellante. We are live in San Francisco for the Red Hat Summit and this is where all the action's happening around open source, cloud, the modern era of infrastructure, software development, dev ops, the whole world's changing a whole new way. We're at an inflection point and we're covering it. Our next guest is a hot startup founder and CTO, Solomon Hikes from Docker, formerly DotCloud. Welcome to theCUBE. Thank you. So you guys are the buzz of the show. Obviously you've got your PR machines in full throttle because we know they hit me up on email all the time but you get senior leadership over there. You guys have been growing. Obviously, Cloud's not new to you guys. So give us a quick update on Docker. What's happening? What's the quick news? We'll get that out of the way and we'll jump into a conversation. Right, so the news today is that Docker and Red Hat are jointly announcing even more collaboration. We had already announced that Docker would be supported as part of the new RHEL version and now we're announcing that Red Hat is committing to supporting Docker in the future. They've announced a gemstone program to train some of their customers to deploying Docker. So there's a whole string of announcements. OpenShift, the OpenShift project has announced that they're standardizing on top of Docker even more, et cetera. Let's talk about Docker for a second. Go back and give us a little history lesson. How many employees you guys have? How much funding you had to name change? Talk about the quick history for the folks who might not know you guys. Sure, so it's 30 of us today and Docker used to be called DotCloud. So we changed names about six months ago from DotCloud. And so we, as part of the name change, we raised $15 million in series B. And before that, as DotCloud, we had raised $10 million in series A in 2011. Why the name changes? Guys fell like, hey, DotCloud was play. You wanted to have a fresher brand. So no one was. You felt it might be fashionable, might be in style. It kind of, we just had to. So what happened is, DotCloud is a platform as a service. It hosts and runs people's applications online, keeps them available, et cetera. And out of that product, as part of that product, we developed a lot of underlying technology, including container deployment and automation technology. And we open sourced that and it became so popular, it became a thing. Even bigger than DotCloud had ever been. And so change in the company was just us admitting, okay, Docker is the biggest thing that we need. So we talk about containers all the time. We always talk about what comes first, the container or the data. And everyone's always argues about containers. It's a really great topic. And first of all, containers are very important. So I want to drill down on that with you. So I tweeted yesterday, do you talk about the container first or data first? Or software first, all three. So let's drill into it. Why are containers so hot right now? And do people worry too much about the container and not about the architecture? How do you balance that? Because you do have to balance it. Yeah, so I think it starts with the application. There's a lot of applications being built today, new kinds of applications, a lot more of them. So I think it does start with the software and what you want the software to do. You want applications that do new things, incredible things. And for that to be possible, you need a new kind of architecture. And so I like to think of the container as the Lego brick that makes that architecture possible. It's the starting point, it's the fundamental unit. Talk about the container a little bit more detail. What is the container? What are you guys talking about? When you say container, define that real quick. So a container is a unit of deployment, right? It's the format in which you package your application, all the files, all the executables, libraries, all the dependencies and one thing that you can move to any server and deploy in a repeatable way. So it's similar to how you would run an iOS app on an iPhone, for example. So Red Hat launched a certification program. You guys, are you shipping it out to Fedora and Linux? Give the details on what's happening with your partnership. So yeah, Docker is now distributed on both RHEL, Fedora and CentOS, now part of the Red Hat ecosystem. And you know, Docker, our goal is for Docker to be available and easy to use on all major operating systems. And obviously Red Hat is right there on top of the list. So Solomon, you know, I've best description, I've heard of what containers do is that really it separates kind of the application management from the infrastructure management. It reminds me a lot of what platform as a service can do also because I should be able to be agnostic to my infrastructure. Can you help reconcile for us as to where containers fit with kind of Red Hat's ecosystem? So, you know, you only work on Linux today. You know, you work across their various Linux distributions, but you know, where does it fit with things like OpenShift and OpenStack? Right. So, well, Docker came out of platform as a service, right? Docker Cloud itself is a platform as a service. And so it's very connected. You know, you could say platform as a service is a specific way to use containers. By definition, when you're doing a pass, you're offering a specialized value add solution to us for certain kinds of applications, right? And we, you know, we were in the business of doing that with Docker Cloud. What we realized is when you want to build a platform that's really universal that everyone can build on top of, you need to break it down into a smaller fundamental unit because there is no single cookie cutter pass that everyone can use, right? There's a lot of customization going on. And so, you know, if you think of OpenShift and other pass platforms out there as, you know, really awesome toys that you can just start playing with the pirate ship, the spaceship, you know, we're coming in and saying, hey, maybe if you build this out of Lego, you're going to be more flexible down the road. So that's the relationship. Great, so, you know, you've also built, you know, quite a good community around your solution. Can you talk about kind of the size of your team and, you know, not just inside the people doing all the coding? You know, my understanding is other than your CEO and Turtle, everybody inside, you know, the company, you know, does coding, but, you know, you've built a lot of external people in the company that are helping to build a solution. And I think it would be more accurate to say the community built us, you know, Docker was a start project and then people showed up and said, we're going to make this a thing, right? And you guys can tag along if you want. I'm like, okay, well, we'll tag along. And so, you know, the community made that project. And so, I mean, I'm the lead maintainer of Docker, but when I wear that hat of lead maintainer, it doesn't matter where I work. Docker, the company provides an infrastructure to support that project, but the project today has core maintainers and all sorts of contributors that are not employees of Docker, the company. And that's, you know, it reminds me of Linux, kind of a smaller version of Linux, and it's a really cool way to work, to do software. Yeah, so is Red Hat contributing at all to that? Or are there any other vendors? Can you talk a little bit about the ecosystem and who's involved? Yeah, so there's, I mean, there are all sorts of participants, many, many rather employees, you know, Red Hat pays for some of their employees to directly contribute to Docker, and that's been a very constructive participation. You know, other companies have participated either directly or indirectly. A lot of individuals, right, if you go through the list, you'll see people who work at, you know, Google, IBM, Intel, you know, places like that, and you know, canonical. Like, now I worry about forgetting names, but there's, it's a big federated engineering effort. So I want to talk about the competition. Obviously, you guys are getting a lot of buzz. Containers do make a lot of sense. It's a great way to have application delivery. You guys talked about that yesterday with Red Hat. You know, application delivery is really a top priority. Getting things out fast, DevOps, culture, you know, pushing code, but people talk about you guys in contrast to vagrant. How do you guys compare and contrast? Some say it's actually complementary. One's lightweight, one's a little bit heavier. Can you compare and contrast the two approaches? Yeah, I think Docker gets compared to a lot of tools out there in the DevOps world, we get compared to vagrant, we get compared even more to things like Puppet and Chef and, you know, configuration management. And I think the answer is the same for all of those. You don't, it's not, you know, Docker is not a direct replacement for any of those. You can use them together. The main thing is Docker just kind of does its own thing. It's a container engine. That's the category. And vagrant is not a container engine. Puppet and Chef are not a container engine. So I think if you use vagrant specifically since you asked about vagrant, if you use vagrant to manage virtual machines, then they are complementary. And the way a lot of people use them together is they use vagrant as a utility to deploy a virtual box or VMware machine as part of your infrastructure. And so on their laptop, it'll be vagrant. On their in-house cluster, it'll be open stack. It might be EC2, you know, in the cloud. And then on top of that, there's a unified layer to abstract it away. And that's Docker. Wouldn't it be easy just to start bickering about who's got what stuff better than new when everything's was all the same? When apples were compared to apples and oranges, we're compared to oranges. But you essentially have a new category. So two questions. One, how did you overcome that? You're a new animal in the marketplace. You're different, but people can compare, contrast you. So I want to talk about your personal experience with how you articulate that. And then two, let's talk about open stack. What does open stack need to do to evolve when you have kind of a lot of competing interests, but not a direct one-to-one match? There are some overlaps, but there's also synergies. Talk about the dynamic. One, you guys as a kind of a new category buster, and also open stack where there's a lot of co-operations going on where it's not necessarily a direct one-to-one. Yeah, I think competition's always good. What's really important though is the value of open sources and standardizing things and federating effort. So the big thing to watch out for at any given time is, do you want to do something new? Or do you want to join an existing effort? And if you're joining an existing effort, are you doing it constructively? Or are you forking and creating kind of a competing implementation? And I think the problem that the open stack community is dealing with is it's really big. There's a lot of interest in it. But it's hard to agree on the single definition of what open stack is. So I think the test is for it to be one thing and to have one name. Every deployment of it, every implementation of it should be interoperable, should be standardized. And that's something we're very careful about. So Solomon, back to Docker. You guys are not yet fully GA. I believe the last version of your product is 0.10. It sounds like we're close. Can you talk about just customer use cases and where you are with kind of customer deployments today? Yeah, so it is version 010 and we release one new version every month. And we've said that it's not yet recommended to use it in production, but we plan on changing that very quickly. The current plan is to make the next release, the first release candidate for 1.0. Meanwhile, a lot of people are just ignoring our advice and rolling it out of production. So production is, it's a gray area. It depends what you're doing, how comfortable you are, getting your hands dirty with the technology. So we have people today rolling on Docker on sometimes thousands of servers running real payloads and we're telling them, hey, it's not off the shelf and they say that's fine, we're early adopters. Yeah, any data points on how many people are doing it in kind of private environment for public environments or you know, is there any sweet spot for it or is it kind of everywhere? I think it would be my instinctive responsibility. It's half and half. And the sweet spot is when you expect things to not be homogenous. And that's a big selling point. If you expect your application stacked to not be homogenous or your infrastructure not to be homogenous, then it makes a lot of sense to buy into Docker on the day one and make that the only constant because it's very small. Yeah, I don't know too many environments that are homogenous. I mean, that's always especially in the enterprise that that's always the problem is it's you know, always heterogeneous and that wipes it off. Yep. So about open source, I was talking about thread there about you were just on was standardization and federation versus corking and going your own. So let me mention OpenStack. So, you know, is OpenStack standard? Are we there yet? How do you feel about that? Because, you know, there was a moment of euphoric hope with OpenStack than this lull of a lot of people jockeying. But we see it swinging back to almost like it feels like it's going to break through but there's still some fun hanging over it. We got OpenStack Summit coming up. We'll have theCUBE there live in Atlanta this year. So it wouldn't happen. We're also going to drill down. So it feels good last year at OpenStack Summit a lot of deployments, a lot of people voting with their code. Yeah. But there's still a lot of jockeying at the past layer. So at the end of the day, it's about software development and getting the apps out there and for the enterprise to feel comfortable looking under the hood. So I mentioned Lego block design. This is kind of like, this is a good sign for that model. So how do you feel about OpenStack? What do you think is the work areas that need to, you know, get snapping quicker and move faster or they're pretty solid on the standardization? So, I mean, first I'm not an OpenStack expert. So, you know, I can speak to my relationship with OpenStack, how we integrate with it. OpenStack has been an important integration point for Docker because a lot of people who deploy Docker as their abstraction layer for applications do that on an infrastructure that is partly OpenStack. So deploying Docker based applications on OpenStack infrastructure needs to be something that works really well. So from our point of view, OpenStack is about infrastructure, right? It's about compute storage. Managing a single plane of glass kind of thing. So, right. And, you know, it's about, hey, I have machines over there. They're managed by OpenStack. I have machines here. They're on EC2 here on Google. Here, that's my laptop. Get Docker work on all of that as a whole. The higher level stuff of OpenStack, it's not exposed to us. We don't use it. And the people who... Well, the developers on the DevOps side, let's ask different questions. So for the DevOps guys out there who are kicking the tires and moving to OpenStack, because the enterprises have demand, I mean, anyone who puts OpenStack on their LinkedIn profile gets like five job offers like instantly. It's hot to be an OpenStack engineer right now. So from a developer standpoint, this distribution, there's a road, there's lighted in the tunnel. So people are making the investment. So what do you guys talk to those guys about? What do you say, hey, Docker can provide for you X? What is it? So, I mean, I think it boils down to what the identity of OpenStack is. People who today deploy OpenStack and use it are people who are deploying and using infrastructure. That's what OpenStack is used for today in the field. And so people who come to us and they say we roll out OpenStack, they have a very clear idea of how they want to combine the two. And so they tell us, I have OpenStack infrastructure, make it work with Docker for application deployment, and we help them do that. So really deploying applications at the end of the day is the number one thing. Yeah, it's deploying applications in a way, in such a way that they are not tied and locked into any kind of infrastructure. So they're portable. So anti-locking. Anti-locking. Anti-locking, yeah, that's right. That's not a good marketing slogan. I don't know. Share the folks out there that are watching, you're in the trenches, you're a tech geek, you're a tech athlete. What is the big deal about this cloud stuff right now? Why is there so much action going on? Why is the developer community, why is there a collision between open source, enterprise, commercial use, and these new paradigms? Why is it a modern era? Explain to the folks in your own words, why is it so important here at the Red Hat Summit? Yeah, I think, I mean, it's a huge trend, right? And the applications that are being built today are being built for a platform that no one can point to. People can point to the iPhone or to Android and say, I'm building an app for that. But the teams and companies that are building, Gmail, Uber, whatever online service we're using, they can't point to a platform and say, that's what I'm developing on. That's my platform, that's my computer, right? It's just out there, it's cobbled together by all these different tools. It's not standardized. It's almost like we're at the same phase for the cloud that personal computer programmers were at in the 70s, right? Everyone has their own, they have a drawer with their custom compiler, their custom memory manager, everything. Custom bootloader and so the trend is that that stack is being built and everyone's in a frenzy to participate in building it for those of us who are engineers first and to make sure, you know, to reap the rewards of it if you're in business. What would you, what's the docker culture out there? You know, the DNA of the company. You guys still are a young company. Get some funding, get some good funding behind you. You're in a hot area, application delivery's hot. What's the culture of docker? We like to build things and we're, typically we're the kinds of engineers who get obsessed about the tools and so all of us typically used to work in an environment where, you know, to build good software, you need to invest some of the engineering time in the tools, right? And some people are a little unbalanced and they get obsessed about the tool more than the initial project and that's the kind of people we are. We just want the tool to be awesome so that if you're building whatever it is that you're building, you have tools that just make you 10 times more productive, right? Just 10 times more innovative. Yeah, so wondering if you could, you know, give advice to people, you know, in the valley, there's so many startups up here. You're a young starter, you know, working with, you know, building a community, getting everybody involved, you know, what have you learned over the last year since you kind of came out of stealth and you know, it's now docker. You've got a big announcement here at Red Hat Summit. You've got docker column, you know, coming up next month, you know, what advice would you give to people out there thinking about hopping on the open source wave and starting a software company? Yeah, I guess, I mean, the first thing that comes to mind is to kind of stick to building what you like to build. You know, docker has been presented as a pivot, but actually it's the closest thing to the initial project we started on 2008. I hate that word pivot by the way, it's such a bad word. You know, we quit our job in 2008 to start an open source project that used containers to automate the deployment of application, you know, abstracting the infrastructure away. That's what we did. No one cared at the time, so we just kind of survived long enough to give it a second try, so it's been six years and it wasn't always easy, but we stuck to it because what else were we going to do? It was the most fun thing to build that we can think of. So I think- Who are your backers? Who is the VCs behind you guys? So, White Combinator backed us in 2010, and then we raised money from Trinity Ventures and Benchmark Capital, and then Greylock recently, you know- Jerry Chen. Yeah, Jerry Chen and Greylock. It's Jerry Chen's first investment, apparently. So, you know- I think so. I think so. Jerry Chen's also a friend of the Cube, so Greylock's some good, you got some good partners behind you. Yeah, they're great. And total funding you raised now? 26 million, I guess. So you got some money in the bank. We did. It's hard to go out of business when you have money in the bank, as I always say, so congratulations, great to see you, and we were looking forward to chatting, obviously Doc, our fan, and Jerry Chen and Greylock and all those VCs. Exciting opportunity, thanks Solomon for coming on the Cube. Founder, we actually have a founder in the house, it's always great to talk to founders of companies, making it happen here. A lot of opportunity, building out, growing, and commercializing open source technologies. All the show rage here at the Red Hat Summit. We'll be right back with our next guest after this short break. Thanks.