 Live from Mountain View, California, it's The Cube at OpenStack Silicon Valley, brought to you by headline sponsor, Mirantis. Here are your hosts, John Furrier and Jeff Frick. Okay, welcome back everyone. We are live in Silicon Valley, California. This is The Cube, OpenStack Silicon Valley. This is the event going on around the cloud, multi-cloud, hybrid cloud, you name it. It's all about beating Amazon. That's what Jeff Frick and I have been talking about. Jeff, our next guest is Madhu Kashyap, Senior Product Manager around OpenStack with Brocade, friend of The Cube. You guys have been great partners and sponsors for us in the past, but great to get you on The Cube here to find out kind of in the trenches what's happening with OpenStack. Obviously networking is a big part of making all that infrastructure work for the app developers. So what's the Brocade angle here at OpenStack? So Brocade is really positioning itself as the next generation of IP network virtualization. So both at the physical and at the virtual layers of networking, Brocade has assets that it's leveraging and building plugins for OpenStack. So both on the switching routing as well as on the network function virtualization pieces of the software that we have. Brocade is building plugins that plug into the OpenStack Neutron project. So Brocade, explain how this fits in with the overall Brocade, because Brocade is a very large provider to a lot of the infrastructure vendors as well as customers. So how does OpenStack did it to all of this? Right, so if you look at Brocade's history and legacy they came from the sand fiber channel space but they've increasingly moved into the IP networking and so as part of the IP networking you're seeing a lot more traction with data center the focus is on the data center networking and OpenStack naturally fits into that ecosystem and whatever that we've been doing in IP networking whether it's switches, routers or the virtual routers, firewall, VPN services that are virtualized are part of that effort to plug into OpenStack and the data center. So give us a quick update on your take from the show here. What is your professional assessment of the OpenStack event here and the community right now? So I've been tracking OpenStack for about a year and a half now trying to do a startup initially before I joined Brocade. And so you can see the momentum and the adoption, the awareness. I used to go to meetups a year and a half ago and you find 25, 30 people. Now there's standing room and all flowing. All these new faces. Yeah, absolutely. But they're really attracting a big community. I mean, if you look at this event, this event really was an ad hoc event sandwiched between Atlanta, OpenStack Atlanta and OpenStack Paris and its packed house. Yes, exactly. Very surprising. Just trying to get even parking for like 400 people here. So, but yeah, Brocade is again, you know, working, we're working with, we have some Lighthouse customers. Yahoo Japan is one customer where they're using OpenStack with our plugins. We did a session with them at the Hong Kong Summit. We demoed some other leadership things that we've taken with MPLS, IPN, IPsec, VPN kind of plugins and NFV plugins. So again, working with a lot of customers who are showing interest in OpenStack and data center networking. What are some of the biggest challenges that you're seeing? Because developers are not infrastructure dudes or gals. They're really more about just making the app and look at the success that containers has had with Docker and just that notion that we had the CEO of Mesosphere on earlier. You're seeing the tooling and some of the software really drive this scale out, horizontally scalable mindset of DevOps and it's really impacting IT. So if that trend continues, that means the app guys are going to be in charge, dictating to the infrastructure, getting the resources and the policy, the policy stuff changes. Which used to be in the networking. So that's flipped upside down. What's your take on this? No, no, so again, very interesting. So, you know, the infrastructure still matters. And so now at Brocade, we're providing all the set of REST APIs that earlier used to be just CLI. And so you have a whole set of APIs that can be programmed. You can program your switch, your router, the virtual router, firewalls, on demand, create VMs. And so Brocade is taking that step to providing the set of APIs so that app developers and other infrastructure DevOps folks incorporate the infrastructure into their configurations and rolling out the cloud and so on. So on the policy side, again, very interesting talk by Martin Casado this morning. And that was a piece that was missing. And I think that's adding a lot of value to the OpenStack ecosystem. So the policy matters. So he talked about an interesting overarching policy framework. Subsystems will be able to absorb that and do their own things, right? So again, I think that is a, and Brocade is very interested in knowing what, you know, how that plays out. Absolutely. So talk a little bit about the challenges of software-defined everything. Right. When it meets real gear and real protocols and real service level agreements and moving data around. So some of the things you guys are doing to address those. So I'll go back to the APIs. APIs are fundamental. And if you saw Adrian Ionel's talk as well, the first thing he had was APIs are critical, right? And so that's what we started with, providing APIs across the board for all of our products. So that will take you to that software-defined data center. You can call these APIs, instantiate links, instantiate flows, instantiate firewalls. All of that is definitely already happening and also the roadmap for expanding that APIs. We're also working. So the neutron plugins that we are working with, that is in, since Havana release, we've been actually working on providing plugins to all of our assets, both physical and virtual. We are taking a pretty big position in Open Daylight as a platinum member as well. So again, we're doing a lot of work, contributing code there as well. So it's all part of the software-defined data center that you're talking about. So we are marching down that path. It's software-driven too. You're mentioning, you guys are involved with a lot of open source projects as well. So talk a little bit about is there, the culture to really adopt, embrace, and I guess embrace is really the word in these open source projects, from what was a really proprietary, you went from proprietary hardware to open source software. It's a big shift. Absolutely, but along our journey, right? Two years ago, we bought Yata, which was a virtual router company and they thrived on open source. So we brought some of that culture into Broke. So we are now cross-pollinated Yata people with the older sort of proprietary legacy sort of mind culture. And so you're seeing the shift in the way people are thinking and approaching problems as a more collaborative, working with the community and so on because now with the acquisition of Yata from two years ago, you can see that shift happening inside of Broke. And are the old dogs getting it? Yeah, absolutely, absolutely. And really the ability to leverage that for a pace of innovation and to drive things. So the market is forcing them, competition is forcing them, so they're getting on. They're getting on? So what's next for Broke? What do you see this trajectory? What vector are you really developing on from a product strategy standpoint? Sure, so we are really embraced open source. So like I said, we are big into open stack, we are contributing upstream, we are working with the community, we are working on framework things, we are taking some leads on things around NFV and we even intubated a project called Service VM or Tacker where you want to keep a pool of virtual machines ready for like firewalls or VPN or load balancers so that they can get instantiate, not in seconds, couple seconds instead of minutes. So that's a project we are taking the lead on. We're taking a pretty big position in open daylight as well. So we're really embracing the open source culture. We don't want to be in proprietary hardware or proprietary software, we want to be able to take that middle ground. So also NFV is interesting because that introduces the fact it's not just enterprise, it's service providers. What are the challenges for service providers with open stack? So with open stack the scale out, the scalability is a big concern, right? There have been different benchmarks around scaling of open stack. You know, people have reached 500 nodes, 800 nodes, 1000 nodes, but you're talking thousands of nodes being able to scale at that level for a service provider. So that is going to be a scale, the AHA capability performance. All of those are still maturing and coming together and they'll have to work through that. But you're great to have you on theCUBE, Brocade. Again, we've interviewed you guys many times, certainly EMC world, all the big events and you guys were also at Atlanta as well. So great to see Brocade out in the open source world. We've had a lot of your folks, you guys have a lot of open source proponents inside the company. Most people don't understand that, but you guys look like a networking vendor. But huge open source culture. Could you explain some of the DNA internally that's driving this open source ethos? Yeah, like I mentioned earlier, so with the acquisition of wired, now we're seeing that culture of open source because that was based on Linux and they built the whole router on the Linux platform. And so a lot of that was acquired into the company, spread through the company, and that DNA was definitely the shift happening in the culture and DNA. Dude, cash, yeah, senior product manager at Brocade here at OpenStack, connecting it all together, open source, the networks got to run, the pipes and the networks got to have B provision and they got to run and be elastic. Thanks for joining in theCUBE. We'll be right back with our next guest after this short break. Okay.