 The Cube at OpenStack Summit Atlanta 2014 is brought to you by Brocade. Say goodbye to the status quo and hello to Brocade. And Red Hat. Here are your hosts, John Furrier and Stu Miniman. Okay, welcome back everyone here live in Atlanta for the OpenStack Summit. This is The Cube, our flagship program. We're going to go out to the events, extract the signal from the noise. I'm John Furrier, the founder of SiliconANGLE. Enjoy my co-student Miniman, analysts at wikibon.org. And our next guest is Dave Meyer, CTO and chief scientist at Brocade. Welcome back to The Cube. It's great to be here with you guys. Good to see you again. So I got to talk to you about what's going on at OpenStack. Obviously, you guys are amazing technology. We had you on The Cube before. You went in depth and geeking out. What are you hearing right now from customers relative to the viability of OpenStack? And where are they in the life cycle of adoption? And that's what everyone wants to know about. You know, our customers are telling us that they really want to go to the OpenStack open source world. That's really, really a major trend and everybody wants to go there. I think that, you know, what's happening too is that it's becoming closer and closer to reality, I think, for everyone. You know, in the early days of OpenStack it was, you know, it was difficult and it was a rapidly maturing open source project. But now, even the tooling around distributions and around deployment is becoming open source. So this is all going to become, the parts of it that aren't already real are going to become real quickly. And I don't know if you guys sat in on John Brasovsky's talk from Comcast yesterday, but he said that they configure all their set-top boxes with OpenStack in the U.S., Xfinity boxes. So you think about that. That's a large, large number of devices that they're configuring. So that's a real deployment and it's at a very large scale. So, you know, there's a lot of real deployment out there right now. What are some of the things going on? We heard some articles about neutron and the network layer and the virtualization is certainly changing the game. There's a lot of action going on the network side. And how does that impact some of the customer pain points around OpenStack? Because OpenStack is a platform. There's a lot of things up the stack, down the stack. You know, at the lower end of the stack, what are the key challenges that you see in pain points with customers? Yeah, so the relationship between the network layer and OpenStack is really shaking out and getting solidifying right now. And being part of the Open Daylight project, you know, we kind of interact with OpenStack on that layer, right? So the question of how much network configuration and things go into neutron versus how much go into a controller layer that sits underneath that. But that is really shaking out right now. And so I think what you're going to see is you're going to see these really robust virtualization technologies in the open source community that are comprised of these layered architectures. OpenStack, Open Daylight. So Dave, we were talking to Neela before about how OpenStack and Open Daylight are coming together. It was good to see OpenStack, you know, opening up to other open source projects. I understand you had a verge of a feather here looking into ODL. I think it was like two hours at the end of the day and you still had, you know, good discussion. So can you give us a little insight as to, you know, what's everybody saying there? With respect to Open Daylight. Yeah, ODL. Yeah, yeah. I mean, you know, the one thing I can say right away is that the OpenStack community is very interested in Open Daylight. In the session today, Kyle Mastry, who's the Neutron PTL, and I had this thing about how, the session about how OpenStack and Open Daylight are co-evolving together. Because Neutron, there's a Neutron plug-in in Open Daylight right now. So, you know, OpenStack and Open Daylight work very well together right now. And, you know, our customers and the communities that we're involved in are really, really excited about this. Because the amount of knowledge, network knowledge and network infrastructure that's in Open Daylight really augments OpenStack. So people are excited about that. So Dave, you know, we're glad to have Brocade here as a sponsor. You know, we've seen you guys hiring a lot of people that are involved in Open Daylight and OpenStack. Can you give us a little insight as to, you know, what you're doing with the team? You know, some of the key people that you've hired and, you know, where are you hiring more? Sure. So when I came to Brocade, I think it was like 16 months ago, I wanted to build an open source team, a real, you know, one of the kind of open source teams that would not only drive innovation into the open source projects, but would really be active in the community. Because I think I talked about this a little bit last time, but the community is really what it's about. And that's where the innovation comes from. And Chris Wright has talked about this a lot, you know, that it's innovation through collaboration through this community. And I wanted my company to be part of that. So, you know, I had some friends, folks I've known for many years through IHGF experiences and things like that, who, you know, were looking around for just that kind of opportunity. Benson Schiller, Tom Nadeau, Colin Dixon, and, you know, it all came together with support from executive level at Brocade. And, you know, we built a, you know, I've seen a couple of things in the press where it's been called a dream team. It's not done yet, but it's, I'm really happy about it. Yeah, you got to be careful. I remember the Eagles came out and talked about their dream team behind the season. It didn't work out so well. So, yeah, I mean, Brocade's had, you know, long leadership and, you know, networking standards. Think about IETF, you know, I know when I, I know some of the guys inside Brocade that are working on, you know, the next speed bump of Ethernet, you know, 400 gig and beyond. So, executive support, you know, is open source. You know, how important is that on Brocade's, you know, top list? It's extremely important. And, you know, one of the things I'm really gratified by is that, you know, those, you know, the company gets that. And that's reflected in the fact that we've hired this crew. And it's not done yet. I mean, we're just, you know, we're just getting going here with this. And I'm looking forward to what we're going to do. Now, we have a pretty good team, and it's come together pretty nicely. But at the end of the day, it's sort of like the, you know, you know, Eagles or whatever. You still have to execute. Yeah. You still have to execute. So, you know, what's new with Brocade and OpenStack this year? What are you guys showing off? How does it fit into the portfolio? What solutions are you driving? And, you know, we'll dig into the partnerships after that, too. Yeah. There were a couple of nice demos done by Brocade folks here. One was integration of OpenStack with VCS, the fabric products, really nice way of configuring VCS through OpenStack, not using Open Daylight. And then there were a couple other really nice demos of the integration of the Viada 5600, which is a software router, Open Daylight and OpenStack. And so we've showed a lot of these kind of solutions. And those kind of solutions are really getting much more robust, much more functional at just an incredible rate. So I expect to see products that are bundled of, you know, all three of those products bundled together in the very near future. Yeah. So we heard AT&T gave a presentation in one of the keynotes talking about NFV. So, you know, where does Viada fit into the whole OpenStack mix then? Yeah. So in the NFV case, so Viada is perfect for NFV, right? Right. It's the NFV set of use cases, which include, like, so, virtual routers, ADXs, load balancers, all of these things. It's the perfect solution for that. And it's also, you know, high-foring performance of software. That's one of the problems that we've been working out in NFV. And NFV is looking towards open source. So that universe of things is really coming together right now. Okay. One of the hot topics of where we kind of need OpenStack to grow has been in the management space. Can you give us your insight as to, you know, where that's going and, you know, where does Brocade sit in that discussion? Yeah, that's a good one. Well, one thing that I noticed this time is that what used to be proprietary management tools, because one of the problems with OpenStack was there was a business that was associated, that people were in, that was associated with the fact that there were so many, there were different OpenStack discos and they were hard to get going. You had management products that would, you know, would address that. And a lot of the management infrastructure is now going into open source. So that will start accelerating as well in the same way that OpenStack and Open Daylight are. Okay. So I can see this acceleration going up and down through the whole stack. So one of the feedbacks we've gotten from people at this show is that Neutron still needs some work to get done. I want to get your viewpoint on that is, you know, how does this get solved? You know, what needs to happen to really, you know, solidify and have, you know, good performance and good scalability and reliability out of networking in an OpenStack environment? Yeah, that's a great question. Neutron's, you know, kind of gelling right now, too. But there's still a, there's still this, I would say, there's diversity of thought there, of course. But the main schism I would say is how much Neutron is going to know about the details of the network. So, and you can think about this sort of how much of a controller are you going to build into Neutron. And that's the real, that's the real battleground right now or the real, the real place where the, at least the architectural thinking is going on. So my sense of that is I think Neutron will eventually, it'll eventually shake out that Neutron will talk to Open Daylight for the detailed fine grained management of the devices and at the Neutron level you'll be doing orchestration. So on the multi-venger thing, a lot of people are saying that it's going to send it around to CIRA. I mean, is that a problem or is that going to change, do you think? You know, I'm not exactly sure what you mean by centered around the CIRA. I think that, you know, Neutron started with NYCRA code. Oh, yeah, okay, yeah, sure. Yeah, sure. Okay, no, I don't think that's going to be a problem because, you know, also as these projects evolve, mature, even though code started, like a lot of the Open Daylight code started in Cisco, right, but it's mature in such a way that that's a distant lineage, you know, and it'll evolve and mature and people will create the kind of infrastructure that they want based on what people are interested in and what they want to do. So the fact that it originated anywhere, I don't think it's really a big problem. So a lot of the developers here are looking for the project and a lot of, we heard from the Rackspace guy earlier, the CTO, John Gates saying, you know, the developer, you can't tell developers what to work on, especially in the open source. You go with the action as nor the value is. What are you seeing for stuff here at the event that's on your radar that may or may not have been on your radar before? Is there anything, any conversations, any developer action happening that gets getting your attention? Oh, there's so many things. You know, one of the main things that's really, I mean, I had this idea before, but the fact that the barrier to entry is so low to, you know, to build a startup or to do whatever you want is effectively zero, right? I mean, you know, you don't need infrastructure to get it on AWS or whatever. You don't need, you can get knowledge on demand because the community is so much about teaching the other members of the community. So not only is the infrastructure not an issue, not only is acquiring the knowledge not an issue, but there's energy, so basically the barrier to entry to innovate in this community has gone to zero. So that's a big thing. And then the second thing is that I've been noticing is that I've had this idea that things are accelerating in a really dramatic way and it's catching that acceleration that people are looking at. So you see a lot of projects and a lot of different kinds of interesting versions of the same ideas in different places because it's easier to do, it's getting progressively easier to do. So that's, those are kind of the two main things. And then the third thing I would say is that, and I've said this in various different ways, but I mean, I compile it down into saying something like what you build isn't as important as how you build it anymore. And that's the same as, I forget who's keynote this was, but he said there's the good, fast, cheap triangle is now the fast, fast, fast triangle. That's the same thing. So it's like all of this is conspired to create the advantage that is all wrapped up in the acceleration and how fast you can go. And that's something we haven't seen. So what's new from Brocade this year and in and around OpenStack? Honestly, you guys have great relationships and talking to some of the partners. You have good reputation and certainly technical company. What are some of the new things around OpenStack that you guys are doing? Well, as I said earlier, you know, there's a lot of integration going on with the VCS product line, integration going on with the MLX product line, including, you know, things like DDoS protection and things like that are all built on top of this infrastructure. And then we're also making a big push into the community itself because that's one of the things I really wanted to have happen. And one of the reasons we built this team the way we did was we want to be community members. And this helps everybody. It helps Brocade because we develop expertise in the core infrastructure that we're building products on top of. And it helps the community because you get developer and committed diversity inside the community. And so those are the kind of things that we're really, really pushing towards. Now on the software product side, the Viata products are all looking at talking Neutron to for orchestration as well. So I think we'll see that coming through all the products. He's got a lot of buzz here too, the HEAT project. Yeah. But you know, okay, so Brocade isn't right on HEAT yet, but we're looking at it. And the thing is, is that Datalog is a language that HEAT wants to use for policy. And the Viata folks already implemented Datalog before HEAT came along. So we're kind of well positioned to work with HEAT. Where's OpenStack headed? From a technical perspective, honestly, you've seen the business momentum. Certainly the hype cycles at the peak right now. You're seeing a lot of customers here, operators, developers. But where is OpenStack heading? In your opinion, where do you see it going? I kind of look at it like a version of the Linux kernel or the kernel. It's core infrastructure. It'll harden, there'll be different distributions like there are now. And that'll shake out over time. And you know, it's like, you don't ask, who do you ask these days? Like what's going on with your NetBSD laptop? Exactly. Well, you don't see that anymore, right? Yeah. So that's what's going to happen with OpenStack. So extracting away some complexity, hardening some things, and then building on top of it. Yeah. And that's the key thing. So what advice would you have for a CIO who's assembled his network architecture and technical staff as they say, all right, let's go do OpenStack? What advice would you give that CIO from a pragmatic technical perspective as they start to really think about laying the foundation for a hybrid cloud or a cloud architecture? What would you, if you could be a consultant to that room of books, what would you say to them? Yeah, that's a really good question. So my sense of the best thing you can do in this community is become part of it. So if I was hiring a dev team, which is what we did, what I would do is I would have them become part of the community first, become core contributors, become committers on these projects, and then help harden the infrastructure and then build products around. So in the case of OpenStack, they have this really nice idea about core versus ecosystem. I think, you know, getting expertise in the core piece of it, just like we're building expertise in the core of Open Daylight is the place to start. It's a community-based thing. I think the power is really there. And then that comes back through not only expertise, not only your visibility in the community and with your customers, but also in your ability to make and develop and produce and maintain products. So that's how becoming part of the open source community in a very deep and serious way I think would be my first advice. So Dave, you know, we talked last time, we talked to you about kind of, you know, the careers of network people. You know, so going back to that, you say, you know, you should get involved in the community. You should do an open source. I mean, it's a significant mind shift for a lot of people. You know, we've always said, you know, if your job today is too much configuring, you know, VLANs and the like, you know, you're going to need to change your job. You know, how's that been maturing? You know, what is, what's the update is? Do you think, you know, network engineers and, you know, the people that, you know, have been updating their CCIEs for all these years should be thinking about? Yeah, you know, it's interesting because I have some friends and they're network engineers, right? And they're always asking me what, you know, what should I do? And I always say, learn how to write code, you know, learn how to write code. And, you know, you don't have to become a developer, you need to, people in that part of the ecosystem need to at least understand what the pieces of this puzzle are for them, you know. And I don't think, you know, I really, I think there's some places where the traditional network engineering will survive over time. But I mean, you know, the automation and the ability to write new kinds of code are the things that are going to be the future because you can't even, you know, imagine a hyperscale data center these days that's not fully automated. There aren't anybody, there's nobody typing on any of that stuff, you know? Yeah, so, you know, Dave, you've watched a lot of the changes in the networking industry and one of the challenges is it takes a long time, you know. You know, how long do you think, you know, it's going to take for this kind of, you know, shift to go, you know, open stack and open daylight, you know, before it's a significant part of kind of the IT landscape? I think it's going to happen pretty quickly. Now, I will say that I think, you know, open stack's obviously quite a bit more mature than open daylight is. So, you know, that kind of stuff is going to happen quickly. And because services are being fielded on open stack right now in various places, public, private, hybrid, cloud and all of that, this stuff is going to happen pretty quickly. So I think, and there's another, by the way, there's another dynamic here is that the cost of hosting facilities in the public cloud, because there's price wars and other kinds of things, is dropping at such a precipitous rate that CIOs actually cannot afford not to do this. Right? Cannot afford not to do this. So the people who are in their network engineering staff are going to have to become facile with this just on an economic basis alone. So on the future of the contribution side of it, I mean, obviously, we were talking with some folks and they say, look at some real metrics around market share. It's not more of a business discussion. But right now, the metric is contributing lines of code. Do you see other metrics coming around the corner as it's mature as other than, you know, lines of code contributed? Because you said it's got a nice model, it's got a core, it's got an ecosystem. The governance looks like it's doing really well with open stack. There's not a lot of mud slinging right now at this point which one of our concerns coming in was, will there be mud slinging or love fest? It seems to be more of a love fest at this point. But what do you see for the contribution model, certain metrics, and can the organic community still win with some of these big leaders coming in like IBM, like HP, like Red Hat doing their thing. So how do they balance that piece? So first one is metrics around success around the community and it has that organic community balance itself with some of the big players. Yeah. So I think that, you know, and we've actually, we've kind of explored the first question in Open Daylight a little bit around what are the success metrics for individuals and for the project itself. And, you know, you mentioned lines of code or, you know, number of commits or whatever. We've been trying to understand if there's a way to enlarge the community so that we can bring more people into sort of the governance and operation of the project itself. So I see, you know, I think that, you know, things like documentation always a problem, things like system tests and integration always a problem. Those people sometimes get left out of the governance, sometimes of the decision making and other things. So I think we'll see an enlargement of that. So that's the first one. And the second question is... The balance between the organic community growth and the big vendors. Yeah. Well, let me take a 1.5 question on that. So if you want to build, so that people just, this is a hard one. Do you want to build a vibrant open source community like you have an open stack or you have an open daylight? You need these big personalities, the people who are not only great coders, not only really understand the open source development methodology and community, but also, you know, are active in the community. So, you know, you have, one of the things I've learned on the TSC is that you have this, you're going to have this interaction between people that has this certain kind of friction, but it's not necessarily bad. Because as long as there's mutual respect between everybody, you know, the development and engineering experience is kind of bad and serial, right? You have an idea, I have an idea, I try to convince you, you try to convince me. And I think that if we can keep that ethic inside open source, it really won't matter if the entities are big corporate players or, you know, they're individuals or they're smaller corporate, smaller players. And we've kind of seen that in open stack right now. Friction with mutual respect is a form of progress. Yeah, yeah. Today I want you to do the last question, share with the audience in your own words. Why is this point in the technology industry so important? You've seen some cycles of innovation. You've seen booms and busts. Why is this point so active? Why are guys our age in the young guns? Why is everyone so excited right now? Wait a minute, guys our age? Whoa, whoa, whoa, whoa. The wrinkles going on here, gray hairs coming in. But DevOps is a young culture, it's a DevOps show. But in general the technology business is changing up and down the stack. There's a radical transformation happening. Why is it, tell the folks out there in your own words what that means, that aren't on inside the ropes, if you will. Yeah, so the reason it's happening in my mind is one is just the rise of software itself. And that took Moore's law in years of these things. But, you know, the thing that's really driving all of this stuff is that the technology we're building is feeding back to cause this. So for example, you couldn't have DevOps, you couldn't have open source or any of that stuff without the internet because you need the seamless communication just for the operation of the projects themselves, much less the technologies that you build, right? So for example, if you didn't have the network, you couldn't really develop software in this way. You may have the network, you have the software being developed, and there's a feedback, there's all these feedback loops inside here. So I think that's the key thing is that there's acceleration in all of these things making them more accessible to more people at an accelerating rate. And hence you get a lot of young people. You know, you can become a developer pretty quickly now because not only because programming languages have become higher level and they're easier to kind of understand, but also because you have the network with all the resources that you need, including live humans and all of that other stuff. You know, I mean, that's what's really going on in my mind is that the technology we built fed back on itself to create the environment that we're seeing now. Dave Marcizio, Chief Sign of Brocade. Brocade, big presents at the show. Congratulations. Good buzz. This is theCUBE. We'll be right back with our next guest after this short break.