 live from the Computer History Museum in the heart of Silicon Valley. It's theCUBE, covering OpenStack Silicon Valley 2016. Brought to you by Morantis. Now, here are your hosts, John Furrier and Lisa Martin. Welcome back, we are live in Silicon Valley for day one of two days of live coverage of OpenStack SV, SV Silicon Valley. 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, my co-host Lisa Martin. Our next guest is Shannon Williams, co-founder of Rancher Labs, hot startup. I've been on theCUBE before, you guys. We've talked to the other co-founder and hot space. You know, go back 18 months. You saw the middle of the fairway docker containers. Now, certainly they're playing through huge. Kubernetes on the doorstep. How does this all fit into what you guys are doing? You know, it's a really good insight because I think 18 months ago when we were getting our company started, we were trying to decide if the market was going to move really quickly or if this might be a sort of a longer burning thing, which as a startup is really important because you've got to figure out how quickly to grow, how much to invest in a team, how quickly to hire people. And there were a lot of really good indicators. We talked to a lot of large companies that were just sort of really impressed at the potential for containers to shake things up. And my co-founders and I, when we started building Rancher, we wanted to build an infrastructure platform that could basically consume, you know, cloud resources from anywhere, virtual machines, physical machines, Amazon, Google, wherever, and implement container management on top of those. And getting started early as the more space was developing allowed us to get our feet dirty with tools like Kubernetes and Mesos when they were just in their earliest stage, you know, contribute, learn about these platforms. And it's really helped us a lot because we started an alpha program back in the fall of 2014 and ran that for about nine months working with a lot of larger, you know, people we'd known from previous lives who were interested in using containers. And very quickly what we found was they were pushing us to move faster. You know, they were saying, we need this stuff now, we've got teams that are already using Docker. We want to put it in production. And so it just spiraled. So I got to ask you, one of the things Lisa and I were talking before, you came on, we're talking about, you know, the Docker madness or someone said Docker sanity. I think Docker madness, we have Cube madness, which is our March madness promotion. And Cube folks know that, but Docker is watching sanity to develop our community, but they don't have a revenue model. And Kubernetes is coming in, OpenStack is shifting, VMware has changed, just saw a rumor come across the grid that Pat Gelsinger might step down. So it's all this, VMware is in turmoil with virtual machines. So are you guys in the middle of all this action? Are you guys in the center of the tornado? Because if you think about like Docker and the revenue model potential of where this is going, it seems that you guys actually might be in the sweet spot, having a nice management platform for containers. You guys see yourself that way? We're gonna open source companies. So it's always, you know, you're always kind of, I think being in the middle of the tornado is always a good place to be as a small startup because you want to get in and talk and work with a lot of customers. And by implementing a platform that supports, you know, Docker's native orchestration swarm, as well as Masos and Kubernetes, we really appeal to a lot of larger organizations that really like the direction of containers are going. But they don't want to necessarily pick one tool to try and use across all the different projects in their company. So what's the lens strategies? You guys look at that and say, hey, you know what? We're not going to have dogma with any one platform. And it's more than just that because you look at a tool like Kubernetes, it's a great tool. It's a really powerful platform. But if you're an organization that's putting it into production, you don't want to be building these Kubernetes clusters over and over again for each team. So delivering, using something like Rancher to build a container service that can support, you know, one click deployment of Kubernetes infrastructure anywhere. And then hand that off to your app development team. That's really appealing. So a lot of the demand we're getting from the enterprise is, you know, my developers want to use containers. I want to have some sanity to that. And a nice open source tool like Rancher that provides a full platform for running containers. It just fits in really well with the strategy. You see yourself as a hedge or almost an insurance policy with a customer. In a way, that sounds what it sounds like. Is that accurate? I think it's more of an enabler, but yeah, I agree. I think there is, you know, one of the problems with any early technology is there's a bit of analysis paralysis. People are afraid to bet on a tool and especially big bets. Always, you know, there's always some leaders who will take those big bets and say, you know, we believe in some tool early on, but usually what you get is companies saying like, you know, team over here is telling us how great Kubernetes is. These guys over here love Swarm. We've got some other team who's been lobbying us to put in a Mesa's cluster for the last year. I don't know which one to do. And so a tool like Rancher that supports them all and sort of says, look, orchestration is, you know, these schedulers are somewhat swappable. They all have value. They have different case probabilities. They're all moving at their own pace. Don't pick one, put in a platform, let people use the tool that works best for them. It's like a light goes off. They're like, yeah, we've used schedulers for a long time. Schedulers by themselves aren't that novel. Schedulers have been part of IT for 20 years. There's probably three today. There might be 10 in five years. Let's just, we'll build out the infrastructure. I can put it before you get at least this question is that Martin Casada had a slide up on the keynote. Developers don't care about the old way Gartner report. They care about, you know, other developers, but the key is try and buy. So your point about these big bets, and that one's, I mean, they're making a big bet in general, but they're not making big bets at once. Yeah, we break the world into, there's two types of projects we see. So it's interesting, we're totally open source. And so very often we don't know somebody's using us until they call us. And so we'll get a call and they'll be like, hey, we really like Rancher. We'd love you guys to come out and do an analysis of our implementation or we want support. But there's two approaches. One of them is it's a team. They've got an app. They've decided they want to run this on Kubernetes. They're using our distribution of Kubernetes and Rancher and they're going live. And they don't care what else is happening in the organization. This app is going live in two months and they've already made their decisions. They're tried, they're using it, it's baked in. They're now just trying. It's a production track. They're just getting sign off by everyone and part of sign off might be having a support contract or having some support. There's another side that is more strategic that's saying we want to build a container platform for our company. We're a Fortune 2000 company and we want to understand containers across the board and set global policy. But those guys, they're actually moving much slower. That's where you get a lot of, well, let's try and do a lot of RFPs. Let's get some analysis. And it's surprising that those projects, not surprising, they'll take a lot longer to code anywhere. But they're both happening. I mean, I see it both ways. I see some organizations that sort of have a lot of central control and they're sort of pushing a central vision and they're going to get there as here's our container platform, use that on whatever clouds you want to. And other ones where they've been given enough rope to run and they're taking tools, picking them and going live. And in a lot of ways, those projects are turning out to be quite successful because as always, if the team believes in the tools they're using, then they'll defend them and make them work. And if they own the decision to choose them, they're a lot more likely to, you know, fight to keep them all working and aligned and upgraded and everything else you need to do. So pros and cons of different models, I think it's going to be a mix, but I definitely agree with the sentiments that it's bottoms up. People try, they test, they use open source tools. They talk to each other in the communities. I'm sorry. Right, no, no, great point. So you've been in business for a couple of years now as open source software. Talk to us about some of the customers, some of the use cases that are really driving the evolution and the maturation of Ranger. Yeah, there's a couple of big, I'd say buckets. One of the things we're seeing is early on in the first year, the people who were really taking containers into production were startups. They were SaaS companies, they were web services companies, they were online gaming companies, and they're just all sorts of types of companies, but they were generally, you know, very, very knowledgeable DevOps teams at small startup B type organizations that were writing cloud native applications and they were moving quickly. And they were just taking our stuff raw, even during the beta, going into production, and we'd find out about it later. But towards the end of last year, we saw a start to see a shift. We saw government agencies, healthcare companies, insurance companies, a lot of telcos, a lot of companies and financial services that were at an application level taking tools like Kubernetes and implementing them with Rancher and allowing their teams or for their own team to put apps in production. So a shift towards more, I wouldn't say necessarily yet we were taking the legacy job application, that's probably next year. This year, what we're seeing a lot of is new applications, microservices driven applications, sort of mode two applications in Gartner speak that are being written and being deployed using containers, predominantly onto VMware and Amazon. I would say that if it's running on-prem, it seems to be running still in VMs running on VMware. If it's being deployed in the cloud, it's going to Amazon, Azure, Google, IBM or other cloud providers that people are consuming. But generally, you're not quite to the point yet. I think if you look down the road, what you'll see more is containers running on bare metal and legacy applications moving to containers. We're already seeing the beginnings of that. A lot of the questions that are coming in right now from users are, how do I run persistent workloads? And surprisingly, it's actually quite straightforward. I don't think, I think a lot of people are, the nature of a container is you have to attach a disk to it if you want to have persistence, but that's not that different than anything else. And so once people kind of understand it, they start moving. And so I expect next year to start seeing a lot of migration of existing workloads into containers to get better density and to get portability. So. All right. Just really quickly, John, thank you. Just one of the things I saw on your website was can running Docker really be that easy? Talk just about the feedback loop that you're getting with your customers to drive the evolution of your software and to really become a leader, maybe a sneaky leader in that space. Yeah, you know, everything in open source comes down to word of mouth. And you know, the first thing you find when you put some software out there and you know, kind of put it up on GitHub as you start to get, people start to find, hey, this doesn't work, that doesn't work. And you know, pretty quickly you get forums up and you start letting people talk to each other. And one of the things we saw really quickly as people started using Rancher, was they liked that it had a really clean interface, that it had support for different orchestration platforms, that it was supporting, plugging into your existing AD and your existing ops tools. And so about three months before we GA'd, we started reaching out to some of the more vocal people in our community and saying, hey, would you mind sharing your stories and sharing a quote and kind of using it, being a case study for us? And I was amazed. I mean, I've been working in software for almost 20 years now. And pulling case studies out of customers is pretty much the hardest thing to do. But in open source, it was just like, I mean, everyone from large insurance, they were like, yeah, we'll put a quote up, we'll write a blog, Sony came on one of our meetups and talked about how the PlayStation group was using containers. And people like Orange gave us quotes for our website. And we've had a ton of companies create their own videos where they talk about how they're using this or blog about it. That's great product validation too. I mean, that's what open source does for it. It is, it's a feedback loop. I think you said it perfectly where they just, they want to share their use case. And one of the cool things we've seen, if you go to our YouTube channel, we've got these videos that are being created where users are just telling their story. But it's not like the old classic VMware, Cisco case study where you bring in a big team. It's like two guys sitting in front of a webcam saying, hey, I'm DevOps Joe one, and this is DevOps Joe two. And we just built our whole CI CD platform on containers. This is how we're doing it. And it's very granular. It's not, I mean, they go through all the problems. Oh, this stuff isn't that good about this, but it's really good here. And so you get a very honest feedback that you just don't get in general and technology. Right, best validation of what you're delivering through the voices of your customers, absolutely. So what's the business model now? Obviously you got great traction on the customer front with the open source. So that's a great testament. Is it just kind of like giving enough cash in the bank? You're not trying to force it? Or is it standard support? How are you guys planning on making money? And not that it's a pressure point, but. No, it is a pressure point actually. It's always a pressure point. Because the reality is, yeah, you have to build a business, I think. And open source is starting, I think there's still a lot of questions in the investment community and in the general community is how are open source companies making money and how will you make money? We've been very focused on building a good open source project to begin with and not taking a closed core or open core closed function approach. So pure open source. Just pure open source being open because what we find is that we get a ton of contributions and we get a lot of attention. What we've been lucky and I think is still very true is when people find value and they start to run real applications on it, we have no problem selling them supported versions of our product that they can count on running at enterprise grade, you know, five, nine level support. And so they want 24 by seven support. They want to know that the people who wrote the software can come and provide, you know. So this is classic open source, great software in production, client will see value in support contract and they're not going to break the bank but enough that's significant. It's not a tip jar but it's big. It's real because for them, I mean the key is they want support. They also want to work with us. I think it's like anything. As you make a tool strategic, you want to have an engagement with the people working on that platform and supporting it. So we found a lot of success in that approach. We're also seeing a big growth among managed services companies who are using Rancher as the basis for their container service. Where they're rolling out a container service to their customer base and Rancher gives them a lot of flexibility. It gives them a multi-tenant platform and we've built a model that allows them to charge for the software as they use it. So that all has been really appealing to companies that are using OpenStack maybe building OpenStack based public clouds or hosted managed services. A lot of VSPP type of companies that want to add containers to the portfolio. And that's another great one. Thanks for coming on theCUBE and sharing the insight. Final word I'll give to you. What's next for you guys? What's your priorities? What are you going to be doing? More open source, more all the time, all the time. What's going on? I think right now we're heading towards, there's a lot of automation that's getting built around this stuff to solve even the more tricky problems. How to do really more complex failover, better storage management. But I think we're still in such an early growth phase. I think the focus is getting people successful and getting them talking about what they're doing. All right Rancher Labs from herding cats to herding cattle. And we've got other new opportunities. Congratulations on your success. Young startup kicking ass, taking names here in the open source community. I'm John Furrier, Lisa Martin. We'll be right back with more after this short break. You're watching theCUBE. Thanks for having us.