 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. Hey, welcome back everyone. We are here live in Silicon Valley for OpenStack Silicon Valley or OpenStack SV. The hashtag on Twitter to follow the conversation is OSSV16. This is Silicon Angle Media's theCUBE. We're a flagship program. We go out to the events and extract the signals and noise. I'm John Furrier with my co-host, Lisa Martin. Our next guest is Matthew Lodge, CEO of WeaveWorks, also former VMware but also on the Linux Foundation, Cloud Native. Cloud Native Computing Foundation. Cloud Native Computing Foundation. CUBE alumni, welcome back, good to see you. Great, thanks. It's good to be back on. So a lot of things going on in your world. Obviously, we'd love to just talk about former VMware and the trajectory of that is. But now with the cloud exploding, hybrid cloud to center of it, nothing more hotter than cloud native, which means basically app-friendly environment that requires a lot of coolness underneath the covers. Like dockers, like Kubernetes, all yet not baked out yet, but getting there. Give us your thoughts on the current landscape of cloud. Well, I think one of the goals of the Cloud Native Computing Foundation is to accelerate the adoption of that style of architecture and that way of building software. I think the big difference here is that this is about applications. So it's really about trying to empower software developers to build faster, better applications, more effective, more resilient. So there's distributed systems design. So it's a little different to some of the previous shifts that would be more about infrastructure. So I got to answer the question because Dave Vellante, who's not, couldn't make it here. We talk about this all the time because it's our seventh year doing theCUBE. In 2010 at VMworld, Paul Moritz laid out the architecture. Which essentially is the cloud native concept now where the moves have been different and awkward at the time but as it evolves, it's playing out almost exactly the way he laid it out. Applications at the top of the stack, playing nicely with some sort of orchestration, middleware layer, and then obviously infrastructure is code, if you will, underneath. So it's kind of all playing out. Do you see it the same way and what's the current version of the map? Well, I think, I mean, Paul is a great visionary. So it's not really a surprise that it's very similar to what he was thinking about in terms of architecture and what he was trying to do with VMware at the time. I think the big thing that wasn't around then was containers and that has turned into this really great enabling technology for this style of architecture. It must have made it a lot easier to make that transition. Because we've been here before in things like service-oriented architecture and similar sort of architectural approaches previously but the execution wasn't really there in the way that it is now. And then from our perspective, we are seeing a big uptake in customers who have chosen to adopt containers and are looking to figure out how the best way to do that. It's not a question of will we do this or we're kicking the ties. It's like we are going to do this. Can you help? So I've got to ask you, this is really interesting conversation because we have a little historical perspective now but this really comes down to architecture and one of the questions I get all the time here on theCUBE and also out in the wild when I'm walking the streets of Palo Alto or Boston or wherever is the number one question is, I want to move to the new world. I want to move my business and be transformative. I want to get the trophy. I want to drive top-line revenue. I want to do better things. And it never seems to be a technology conversation per se. Here's some stuff to work out but it's an architectural mindset shift. Can you share with the folks that shift and what that preferred architecture is? And kind of take a minute to explain that because I think that seems to be the one thing that people seem to be asking for. Yes, the thing that people are after is agility. They want to go faster. That's the business driver for all of this and that's why they care about it if they're on the business side and not on the technology side. And so if you're a hotel company, in the past you've had a booking system, you had an IT team but you probably bought something off the shelf and customized it. You didn't necessarily write any software yourself. Today you're trying to compete with Airbnb who has a really slick application and a nice website. And so writing software yourself is now part of competitive differentiation. It is part of the value your customers perceive of you. And so- First outsourcing some contract firm or- Yeah, or taking something off the shelf. Right? And you're looking to be different. General purpose software. Right. And so that has changed. So a lot more organizations care about writing software than ever before. And then they want to write software the way that Facebook and Google and Amazon and all of those guys do it, Netflix, all those guys. And so Netflix has really popularized a lot of the sort of architectural approaches and the patterns here. And containers makes it much, much easier to follow that same model. You know, it was quite difficult before. You know, Netflix was way ahead of everybody else. Containers make it much, much easier. And so that style of architecture lends itself very well. And that's why in the charter of the Cloud Native Computing Foundation we're talking about it being container-based. Because that is the substrate that all the stuff is running on. So that's what the business people are all about. So let's break down that architecture. I know agile is not necessarily an architect. It's a concept and methodology. What architecture specifically containers would be in an architectural component? What are the other things that hang around that? How does it hang together? So they're breaking down into microservices. The idea that you can take an application decompose it into these individual pieces rather than having lots of people working on one big code base, having lots of small code bases. And when you do that, lots of things change. Right, in the past, you had one code base so you'd pick one language for your application. Because everybody had to write on the same code base, right? So you couldn't have people write in different languages. Well now in microservices you can. If there's some part of your application that needs a functional language and that's a better way to solve that problem, you can write a service in a functional language and it doesn't matter. Pick the tool for the job. Pick the tool for the job. And make it a microservice. Same with databases. You had to all agree, you end up with Oracle database because that's got all these features you need to do everything in your application. Well now maybe there's something better for time series data. And there's a different database for unstructured data and then maybe there's Oracle for your relational stuff and transactional integrity. And so it's really changing the way that software is being written. So speaking of change, one of the things that was talked about earlier today and a number of times in the program is the cultural change that's needed to enable companies to become more competitive. How are you seeing enterprises with traditional enterprise applications trying to manage that and embrace cloud native applications? How are you seeing them do that successfully? So what we're seeing is that they pick a project where they think something in cloud native and container-based is going to help them be successful. And so they pick something that is important but is not sort of bet the farm. And they put a team together and what we see is that that team works together and they're trying to figure out what is their new stack going to be, right? And how are they going to orchestrate this application? How are they going to architect it? How are they going to organize? And then they're going after that and building it out and learning from that experience and then picking the next project to maybe something a little harder. So out of customer last week, they're going to start with all of their stateless services for their website, right? So all the things, they're going to avoid some of the hard problems around persistent data and some of the other challenges and start with all the stateless stuff. And they're going to do that first, they're going to learn from that experience and go from there. So the iterative approach, picking something that's important but not too important seems to be pretty successful. And so that's helping them to be able to launch things faster to be more competitive or is that in some cases an inhibitor? It's helping them to launch things faster and be more competitive. And it's also helping them build applications that are less fragile. So there's the expectation of 24 by seven availability for everything, right? So they need to be able to do that. In terms of growth, wanted to pick your brain on some of the collaboration that we've has done within this community, this open source community. How are you leveraging the community to grow and really seize new opportunities for Weave? So, as an open source company, it really is all about community in terms of how you build the product in the first place, the interaction you get from open source but also being able to work with other open source components. And that's one of the things the CNCF is keen to do is incubate different open source projects. So for example, the second project into the CNCF is Prometheus. So we're adopting Prometheus as our monitoring and time series and all of the data for Weaveworks products. And so being able to collaborate that way and work with other organizations also interested in similar challenges. How do you monitor distributed microservices based application? Open source has really emerged as the way to manage that kind of collaboration. Tell us about the open source foundations and the community here because we've been following OpenStacks there from the beginning. Of course, I was very critical of the foundation it felt like a marketing program but they quickly changed and became very code oriented. Remember the second, third year was vote with your code, changed at the top and then really since then there's been a great trajectory really driven by I'd say hardcore open source coders but yet an eye on the prize. They've zigged and zagged, dodged some bullets if you will. Certainly the growth on Amazon is just spectacular. As yours come in, elbows fly in. Yet it's healthy and it's survived and on the peak of thriving. Right. What's your thoughts on that next step? What needs to happen for OpenStack and OpenCloud to thrive? I think it's really at a turning point. I think the Morantis announcement under Craig McClucky and I think Brandon from Coro is talking about this this morning here at the conference. And the realization that it's about this shift onto this next model of architecture that we've been talking about here. And so the question for OpenStack is what is OpenStack's role in that environment? Because you could say, well I can run I can run all my stuff on containers now. So why do I need to run VMs? And I was talking to a technical architect or one of the bigger players and so they have customers like that now where they have all these very long lived VMs running on these OpenStack clouds and they're running Docker. So there is a lot of churn and a lot of software development and lots and lots of change going on that's happening at the container level and not at the VM level. The convenience of containers is quite interesting disrupting the VMware market or virtualization. Well anybody who's focused on the VM right it's a challenge for you. So it's true for VM over it's also true for OpenStack. So the question for OpenStack is okay if I'm thinking about my next, what do I roll out? Do I roll out Kubernetes or do I roll out OpenStack? Do I have limited balance on the team? Is that what's the right thing to do here? And so figuring out what OpenStack becomes in the future I think is a key question for that community. And you think that they're in a good position to solve that problem? Well the interesting thing about OpenStack is it's gotten broader, right? So it's not just about running VMs. You've got identity services and your storage and your different flavors of storage. You've got all kinds of different projects. You've got things for running, managing bare metal that are still very relevant. So you've got all these different components but maybe in the future OpenStack looks different really just for that. Sorry John, one of the things that they were talking about in that context this morning was this need to, it's not us versus VMs containers bare metal. It's and with. So looking at some of these larger enterprise organizations is your vision that they're going to have to coexist very well OpenStack with these other players in order for them to succeed just as these enterprises were dependent on OpenStack to be competitive? Well I think if you've already deployed OpenStack it definitely is and. The question if you haven't though it is an either or decision because OpenStack is a big, complex thing to deploy. And that's always been the criticism of OpenStack and it's why you've got really great services with people like Mirantis and IBM and Red Hat and who've got it down to an art for like we're going to make this easier for you. But the question for an organization is if I'm really trying to serve my developers what do I really want to focus on? What do I want to spend my time and my resources and that adoption? And so I do have to make choices. Matthew, talk about what you're working on now. What are you excited about? The Cloud Native Foundation certainly is fun. But you also are chief operating officer of a company. Weaveworks, I see surprisingly, not surprisingly microservices and containers. So hence a little bit of bias but I think it's right on the money. I don't think it's bias in the sense of over bias. Those are the people we talk to, yes. I mean that's, we are seeing the same thing. But what are you guys working on? How's business? Give us an update on what Weaveworks is doing and what you're excited about. So Yeah, Weaveworks is primarily known and the first thing we did was networking for containers and making it very easy to deploy your application across multiple hosts, multiple Docker hosts and making that very simple. And then we graduated from there into understanding, okay, I rolled out my application, I wired everything together and that was into doing and that's WeaveScope was our second major product and that's going very well. So for us, you've seen a lot of adoption of WeaveNet, it's a crucial challenge that many customers need to solve and then WeaveScope is picking up as people realize, okay, now I built this thing and I'm having a hard time really understanding what's going on and I have different problems now with microservices than I had with applications in the past in VMs. Because I could go look at the VM and I could look at CPU and memory and IO and I got a pretty good idea of what's going on. It's not like that with a microservice. So would you agree that automation which is everyone strives to get to, obviously that's pretty, not well, doesn't need to debate that but getting to automation, the challenge is you have to see it and understand it before you can automate it, right? It's kind of like common sense if you think about it. Do you agree with that statement? I think that's true, but I think even folks who have automated, they're like, well, is it actually doing what I think it's doing? And we saw this ourselves, we were developing this sock shop example demo application for microservices and we saw this container like appearing and disappearing in the live view and we're like, what's going on there until there was a bug and it was actually crashing and restarting. You automated the wrong task. It was great. Scaled up the bug. It ran most of the time. It said it would crash and restart. And so we didn't really notice because it kept working because it kept restarting, but that's the problem. Plus the problem with scale, right? The scale is you scale what you have, right? If it's good, it scales great. If it's got a little tweak in it, it could go off skew out, right? So that's a problem. So management software is a big part of what we're seeing. State of the Union there real quick, where we are with that, what's interesting, what we saw with management tooling is when AWS came along, it opened up opportunity for new players, right? And you saw a new relic emerge from that as a really strong management offering designed for cloud, a cloud infrastructure, a cloud AES. So we think the same thing is happening again with microservices because it's not just about running VMs now, it's also, it's about understanding the aggregate behavior of all your containers in a microservice and sort of being able to connect the dots for all those. Matt, thanks so much for coming on theCUBE. Appreciate the insight. Great to talk microservices, orchestration, all this great stuff. Thriving, surviving and thriving, that's the theme here at OpenStack Summit, continuing to mystify the skeptics who have predicted its death years ago. So congratulations on your success at the cloud native foundation and we'll be following you there. This is theCUBE, I'm John Furrier, Lisa Martin live in Silicon Valley. We'll be right back with more signal from the noise after this short break.