 Live from Vancouver, Canada, it's theCUBE. At OpenStack Summit Vancouver 2015. Brought to you by headline sponsors EMC and jointly by Red Hat and Cisco. With additional sponsorship by Brocade and HP. And we're back. I'm Stu Miniman with wikibun.com. Here with SiliconANGLE TVs, live continuous program from OpenStack 2015. Beautiful Vancouver. We go out to all the big enterprise shows, help extract the signal from the noise. And one of the key themes we've been digging on this week is really talking about the stability of the code, the maturity of the code. And networking's really at the center of that. Joining me for this segment is Ken Ross, who's the director of product management with Brocade. Ken, thanks for joining me. Yeah, thanks for inviting me. All right, so Ken, your first time on the program. Can you tell our audience just a little bit about, you know, you, your role at Brocade? Yeah, so from a product management perspective, I look after the network management. We have a number of cloud orchestration initiatives as well. So that gives me a good perspective across the whole company. And I think what we're doing with OpenStack is getting really interesting. Yeah, so Ken, you know, I've worked with Brocade for a long time, for 15 years. We were talking off camera a little bit is, people when they think of a company often aren't thinking about, you know, oh, here's the cool new thing that you did today. They remember like, oh, you know, it's, you know, I remember, you know, 15 years ago, when Fiber Channel was rolling out and doing that. So, you know, Brocade's got, you know, leadership in the Fiber Channel market, but, you know, really a transformation in the company. You know, very big presence here at OpenStack and a lot of software people. Ken, can you talk to me a little bit about, you span kind of some of the old and the new. How does that play out? Yeah, I think you're right. A lot of people think of Brocade as the Fiber Channel business, but really we've put a lot of investment into the IP side. So a number of initiatives that we've got going with our virtual router, the NFV space, and then also on the IP fabric space as well. And that's actually starting to get some good traction. So really a lot of our emphasis, particularly when we're dealing with OpenStack, is in addition to the Fiber Channel piece, looking at what can we do from a plug-in perspective, covering both our fabrics on the IP side, as well as the software-defined networking function and what's going on with that telcospace. All right, Ken, can you give us a little bit of a high level, you know, where does Brocade contribute to OpenStack? You know, any speeds and speeds, how many people, how much code, and the like? Yeah, you know, we've spent a, probably the last couple of years involved in OpenStack. We started with our IP fabrics and the Fiber Channel fabrics. We've been broadening that out now, particularly with the V-Router integration. So a lot of initial work around plugins to cover all of our product lines. And with the Keto release that's just come out, we've now basically got that in place across all the product lines. Now in addition, we've been contributing from a community perspective, really as you'd expect on the networking side, into Neutron, where we see some issues that need to get fixed that tie in with our product lines. We've also contributed in the Fiber Channel sand zoning piece, and that's started to pick up with our OEM partners. So one of the things that we're doing here this week is really starting to dig into what other community projects are worth us, really starting to look in and try and further the progress, things like Congress, again, kind of focused around the Neutron space. Okay, yeah, we had a good conversation with Dave Meyer and Tom Nadeau talking about how ODL and Neutron are going together, and of course Brocade had some key leadership there. So Brocade had the Viata acquisition and you've grown out really, not just the IP side, but the software side of the business. The software side is really interesting because we've got both the tie in with ODL, really good compliment to what's going on within Neutron on the OpenStack side, and then tying that in with, what are we doing with virtual routing, some of the more recent acquisitions with SteelApp on the load balancing side as well. So we've been involved in Neutron on the load balancing side and we're looking now at how do we tie in the SteelApp piece to that. Yeah, all right, so maybe talk to us a little bit about kind of customers. What are you hearing from customers about OpenStack? Is there really a pull from them there, or is it just kind of coming through and saying, hey, we're doing that OpenStack thing if you want it? Yeah, yeah. You know, I think the last two to three years ago, we did a lot of education of the customers. They were interested in what was going on with OpenStack. They could see some building momentum, but from our perspective, it was more us educating them. And then probably 18 months ago, we started going through a phase where a lot of the customers, and this is a networking perspective, right? So a lot of customers started off with OpenStack on the compute side. They weren't too worried about the networking side yet, particularly when they were in proof of concepts. I would say it's been really interesting over the last six months or so where the conversation has really changed. So a lot more interest, particularly from the cloud service provider space, the operator space, where these guys are obviously bidding their infrastructure, infrastructure as a service type offerings, their business on that. They need to sort out the networking piece. And you can tell that from a customer benefit perspective, as they're looking for that agility, if that agility is now translating into an SLA that they have to meet, then the conversation with us has changed. It's changed into not just how are you, what kind of integrations do you have, but how are you doing it? They're really concerned about the robustness of the integrations and the plugins. Because we have really good quality APIs with some of the Yang models that were maybe talked about yesterday with Tom and Dave, that is a robust interface to go and do that integration. And that's really helped the confidence on the customer side. So I would say in the last 16 months, a very obvious buildup in actual deployments and planning that is tying in with what these guys need to do on the networking side. Yeah, it's interesting. We have a lot of discussion at the show here about really kind of the enterprise user that might be consuming from those service fighters, companies like Rackspace, HP Healian, many of your OEMs are in this space building, Clouds. We know in the networking space you kind of span enterprise service provider and even kind of the web hyperscale guys there. From a product management standpoint, how do you kind of span the requirements from the traditional enterprise out to some of those much bigger guys? Yeah, it can be a challenge. I think one of the things that we're really looking to do going forward is focus in on that cloud service provider space, get that infrastructure as a service piece sorted out, prove it out with the cloud service provider guys. I think one of the things that's been good in terms of having some of the initial deployments is that we've actually proven and hardened some of the integrations that we've done. So for example, cloud service providers offering a service, and some of this is internal to their own organization, offering a service to their app developers and not having to take that service down while they're still building out the infrastructure underneath. So they're now in a position to come along, build out additional switches without taking down the service from the OpenStack perspective. We've put buffering into the plugins and it really helps that stability and the confidence in that. So we've had some good experience stabilizing that. That helps with the adoption from additional customers. Yeah, it's interesting. We've been having a lot of conversations internally from a research standpoint is there's some technologies that the typical enterprise probably shouldn't even be bothered with. OpenStack tends to be one that enterprises might do but if you look at the whole SDN discussion there, some of it, I want to be able to go to somebody that can handle that challenge for me rather than going through some of the pain ourselves. How do you look at that? You know, we'll see some of the enterprise customers where they're particularly tech savvy and they have their own resources, they can handle it, but that doesn't really apply to a lot of them. So really what we're doing here is looking to partner. We don't do a distribution ourselves. We do certifications with both Marentis and Red Hat and point the enterprise customers to those guys. That really helps and we'll be backed off the support side as well. So they get good quality support. Both Red Hat and Marentis can bring in the expertise on the OpenStack side to get the projects going. That backs off into us if there's any issues and so we're really being focusing on those certifications. Yeah, so one of the big news from the event here has been talking about kind of that big tent and death core. Does that make your job easier to be able to kind of identify, you know, okay, here's what is actually going to be an OpenStack, how we can get, you know, powered by OpenStack certified? You know, does it help with that? A little bit, we're obviously keeping an eye on what that, the implications of that on the networking side. So then that ties more into what these partners are doing. I think it's interesting what Marentis have announced earlier this week with OpenStack unlocked and that partnership and that set of certifications, that's really going to help. And also I think the ecosystem is starting to get to a critical mass where we'll see more customers adopting it. Yeah, so, I mean, we mentioned, you know, Brocade's not looking to have their own distribution. You're a partner with companies there. Do you think there, is there a big market for the distribution because if you look, there's companies like, you know, there's Cisco InnerCloud, who's probably not going to be your top partner, but you've got, you know, Red Hat with their distribution, Marentis who's made a lot of partnership announcements, you know, HP Healion with what they're doing. You know, how much of it, you know, where do you see the distribution fitting into all those? Well, I think it helps. I think it's going to be interesting to watch how much adoption there is actually of these cloud offerings as opposed to a separate distribution, but you've got a tension going on between the companies that want to build their own private cloud and then the public cloud. How do you do that hybrid piece? And I think there's a lot of focus going into that just now. So as an example, again, an announcement from EMC over at EMC World and then again this week with a tie-in to what we've been doing with our reference architecture with them. So that reference architecture then helps both the cloud service providers and the enterprises to go and adopt. And I think that hybrid piece is something that is going to get a lot of attention over the next 18 months and we'll see how that matures. Yeah, absolutely. So Ken, can you give us a little bit of a forward-looking, you know, where's Brocade's efforts with OpenStack heading? Yeah, so a couple of, I mean, one of the things we're doing this week is just digging into one of the most obvious areas to go and invest in. I think one of the drivers that we're getting from those cloud service providers is really looking at how do we start taking this to the next level in terms of scale. So how do we scale out? We've done some hardening in that area. We're starting to look at extensions across data centers. So we've got a involvement here with the community on the VXLan gateway work. We're internally, we're kind of looking at how do we expand that to leveraging VXLans in a layer three fabric across data centers. That's an area where we are expecting to do a lot of proof of concepts over the next six months. We'd like to see that brought back into the community and I think that's one of the key areas. There's a number of others tied in with the NIV space. What we were talking about earlier on with the virtual router, looking at partnerships there and what do we need to do tied in with the architectures coming out of Etsy with the Manus specs. And then I think lastly, increasing or building on what we've done in the storage space into customers that are looking at dedicated IP storage networks and how does that tie in with OpenStack? So we've got a couple of guys here just digging in as part of the design summit looking into that space as well. So I have to ask you about the containers piece because in the keynote this morning, where's the future? Well, we know containers is going to be a big piece of it. Networking's been a bit of an open question. Docker made a move to buy Socket Plane. If you look at CoreOS, one of the cool projects that I think is amazing that they did is called SCD, which in many ways, I think it's going to give some of the value that we've been talking about from SDN to be able to manage multiple devices, be able to have them kind of manage separately. What's Broke's take on the container space? Are you guys partnering with kind of the Dockers and CoreOS's of the world? So we're really taking a look at this. I think one of the directions that the networking side is going to take is looking at how you can actually run containers on the network. That opens up some new applications. The speed at which you can actually do that will definitely help in terms of the difference between that and the VMs. An area that we're really just starting to look and explore into. Yeah, absolutely. We know no matter what we add with that it's going to put a stress on the network. I mean, there's always more data pumping over the traffic, always more, you know, if we talk smaller pieces and changing often in moving places, I mean, V-motion caused some stress, you know, we kind of said that if there was a VM blender was a problem, you know, the container piece is probably going to be in order of magnitude more. Yeah, it's interesting. I mean, we've already seen over the last four to five years the VM side of things leading to some transient congestion issues in the network. We've put instrumentation in there to capture that. So I think we're in a good position in terms of what will happen with the containers and just becoming much more dynamic. We just continue with that. I think it's going to be interesting to see how we can leverage the container technology on the network itself. Yeah, I mean, Ken, one of the things we say on theCUBE here is, you know, we're fortunate to live in interesting times, you know. Is OpenStack, do you believe this is really an opportunity to kind of, you know, push forward some change in the industry from the networking standpoint? Yeah, absolutely. I think the networking industry has been stuck in the dark ages for the last 20, 30 years. We've got CLI scripting as the method de jour for doing operations. We need to move that. So we've actually started taking a close look at how do we leverage some of the technologies from a DevOps perspective into network operations. There was an interesting session yesterday from Rackspace about how they're doing the build, release, deploy process for OpenStack itself. We're looking at how do we take that into network operations. That's a new set of tools for the network engineers. We've actually had made some recent hires of network engineers who really understand this space. They've started applying these kind of concepts to networking, and now we're going to look at how do we help move the industry forwards from that perspective, and that dovetails very nicely into the SDN space and the controller. So you actually have to tie in both the initial physical build out as well as what we're doing on the controller side. All right, so Ken, we were talking off camera. You haven't had a chance to really enjoy some of the beautiful sites in Vancouver. No, I haven't. Tell us, you know, what's been impressing you here at the show, what have you seen, and kind of share a little bit of flavor from OpenStack Summit? You know, I think, you know, being going to the show here for the last two or three years, each time there's more and more people, just looking at marketplace last night, we could hardly move. There was so many people in the booths, so really good set of momentum. I think a lot of more involvement from the marketing side, which helps, I think, the cause. We have to have that tied into actual deployments, and I think, as I said earlier, this is the last six months where we've really seen that stun to pick up, makes a big difference. We can see our, the engineers are really happy that they've seen code that they've deployed into, or upstreamed into the community starting to get used, so exciting times. Yeah, it's the truth. The booth crawl last night, you know, there's so many shows you go to, and it's like, all right, everybody lines up for the beer and then kind of talks on the side. There are a lot of really good discussions talking, you know, with the beer flow. Yeah, there was. Yeah, it was interesting actually, even just talking to some of the channel guys now, they're actually gearing up with the right kind of resources to start helping us do some of these deployments. Oh, boy. So, yeah, you know, you bring up, you know, talk about the channel, because you know, one of the criticisms we've had in the industry is, you know, it's a little bit slow to adopt certain people. If I look kind of in North America, you know, I'd say I'm generous to say 10 to 15% kind of really understand how to leverage cloud. You know, do they have a service to be able to help justify, you know, what do I do with AWS or Azure or service providers and the like? You know, talking about the infrastructure guys that might sell your switches, sell storage, sell them to you. Yeah, yeah, I mean, I think generally that is true. I think what we're starting to see is some leading indicators of some of the more forward thinkers in the channel who are actually starting to invest in this. So it'll take some time, but they're starting to gear up with the right skillsets which are pretty difficult to get your hands on. And I think that will really help. So it's a leading indicator. It's still got a long way to go. All right, so, Ken, we've talked about some of the good progress here. You know, what more do we need to do? Where's Brocade focusing to help move, you know, OpenStack, you know, really kind of across that chasm into more maturity? Well, I think we need, we're beefing up what we do, these certifications that we do with the partners. I think this is going across both our OEM partners. So we're working pretty closely with HP, EMC and IBM. We tie in with Red Hat and the Marantis guys. We've got a much broader portfolio now that needs to go through those certifications to make it easier on the consumption side. And then we're beefing up, as I said, both the professional services and some of those partners to help get that pushed out in the field. So a lot of investment, really more from a go-to-market and deployment side where it is up until now it's been more on the engineering side. All right, well, Ken Ross with Brocade, really appreciate you coming on theCUBE, sharing with us your experience of the last couple of years and how far we've come and the next step we need to take. We'll be right back with lots more coverage here from OpenStack here in beautiful Vancouver. Thanks for watching.