 Hi, I'm John Furrier with SiliconANGLE.tv and we're here inside the queue with Nick Vandersby. Welcome to Inside the Cube again for the second time. You are here at HP Discover and I'm here with my co-host. I'm Dave Vellante from wikibond.org and John, excited to be here again. Yeah, it's the last day we're kind of like kind of getting our wind here. We're in the mobile studio outside the big keynote and Paul Moritz gave the vision about what's going on with virtualization and VM and moving machines around federation, multiple clouds, running everything on a private cloud and Nick, we had a conversation at HP Discover about cloud and one of the things that we're seeing here, we want to talk about is the impact of the whole cloud pressure points on other parts of the network. Obviously, VMWare, we're hearing about what's going on in the middle of the stack with software frameworks and all the things that are doing around virtual machines, VMotion, et cetera. And then you got the applications we heard from the CEO, socialcast and you know, Zimbr was out and been out there. So, at the bottom of the layer is the infrastructure. So, share with us, just your perspective on cloud because you were on the cloud side. You're pushing that strategy and business forward. It affects other areas, just talk about your view and what are those pressure points that we're seeing in the market? Especially in the cloud, the cloud is driving the network to get a lot more malleable and changeable and fungible on the fly because if you're going to provision resources and move resources around, you need to be able to have a network that's changeable on the fly as well. I know with what we've done with an offering that we call HB Cloud System, we provide a turnkey, private cloud or public cloud to a service provider and with our cloud, you cannot just deploy virtual machines but you can deploy physical machines into it as well. So, maybe I want to deploy an entire Microsoft Exchange environment. There's multiple virtual machines, multiple physical machines. Well, in order to do that, we've enhanced the fabric through something that we call virtual connect and we can dynamically change a server to have on the fly two NICs and two HPAs or reconfigure it and it becomes six NICs and two HPAs which is what VMware wants if you want to lay down ESX on top of it but if you want to run an Oracle database on it in the next minute, it's got two NICs and two HPAs so the network is much more fungible and changeable through software, through software control as opposed to static NIC, static design for the environment. So, virtual connect is able to do this on the fly for us. Yes, so we went through this kind of network exercise backed in the consolidation days of the late 90s, early 2000s, the original sand days, you consolidate stuff and it was a real networking exercise, almost a science project, right? So, are you seeing that times 10 and is the industry responding in a way that can actually meet that sort of vision that you set forth? Well, we've actually simplified it to tremendously for customers instead of it being X times 10, it's much more simple because you can take any server as a minimum of 20 gig of capacity bandwidth on it and you can connect to any network, to iSCSI, to Fiber Channel, to Ethernet connections and slice and dice that up at will on the fly. So instead of having to pre-think things months ahead and order it and put it in place, you just get an HP Blade system with virtual connect, you want bandwidth for your database, you got it. If you want it to reconfigure it to be an ESX server, you've done it and- It takes the manageability nightmare of the equation. And the cost as well, I look at just for instance, if you put in place 16 rack mount servers in a traditional way, hooking up VMware into that environment, VMware would tell you they want six SNICs and two HBAs in every one of those servers, one for VMotion, one for data, one for administration, times two, because you went backups and then HBAs for storage. So that's, you're buying 217 network parts, cables, SFPs, NICs, HBAs, top of racks, switches. And what we've done is we've simplified it, you buy HP Blade system, you buy two virtual connect modules, you're done. You have the equivalent, you can have six NICs and HBAs on all 16 of those blades or reconfigure it any way you want. So it's 95% less complex, because there's 95% less parts, two parts versus 217, and costs you 65% less, like $185,000 less per rack of infrastructure of servers when you use virtual connect comparative. So let's talk about performance. And you're able to accommodate 10X, my point of the traffic, the complexity, the data of that old world is really what's going on. And much more because you've got tremendous bandwidth. With some headroom. We can juice up a single blade to 192 gigabits per second. Juice up a blade, I like that. Juice up a blade, you guys. Juice it up. So that's a lot of bandwidth that you can slice and dice any way you want. And so at a minimum of 20, our biggest blade, 192 gigabits of bandwidth and two terabytes of memory to go along with it. Now that's a VMware monster with just that one blade. So what are the challenges? I mean, obviously you have all kinds of new performance levels that you're seeing. We heard Mike Banek say, obviously performance is the number one thing. But it's kind of like, it's always been talked about the zero sum game and networking and computing. It shifts somewhere else. In the cloud side, where is the cloud weak? Obviously the provisioning and all the virtual connect stuff is fantastic and enables a big campus to do cloud and other companies. So what's going on in the cloud level now on top of that? You're kind of right. Because it kind of pushes the envelope because when you solve one problem, the next one shows up, right? So with cloud, for instance, if I provide you with a cloud system for infrastructure as a service and I can divvy out virtual machines that will be a self-service portal or even an entire, an exchange environment, multi-tier, multi-node. But if all you've handed to that consumer, that person in the requester, is a bunch of VMs and servers running OSs, the next person down the line, the application person is now the bottom, right? Because you've brought up all the infrastructure but the application isn't installed. So instead of yelling at the server guy in the storage guy, it's done with the push of a button. Now somebody's got to deploy the app, the software and configure it and do compliance. So infrastructure as a service, you start getting into software management, compliance, patching, updates, platform as a service. So that's why we've got that offering that starts with infrastructure as a service, cloud system matrix, moves up into platform with cloud system enterprise because you're 100% right, solve the problem here. Now the application guy is the bad person because he's the bottom. So Alex Williams just wrote a post up on siliconangle.com where we have the reference point for tech innovation. He said a post which we were talking about earlier is the problems are here today. Everyone's talking about the future, but the problems are here today. What are those problems? You can talk about cloud burst and something that's passionate, close to your heart that you're passionate about. So is that addressing a current set of problems is it future-proofing the headroom that's going to come out of the existing room? So just break it down and where we're at on what's on the table in the marketplace today. Well, what scares the daylight out of you. The bejesus out of every customer that I talk to is, okay Nick, I've got, I said a traditional infrastructure that's not going away anytime soon. I think I want to build out a private cloud and there's these public clouds out there that I can get my resources from as well. How am I going to be able to handle this? I'm dealing with sprawl today in my traditional infrastructure. Is my private cloud going to be another management system and a whole other set of stuff to manage? And my public cloud is a whole other stuff in a new management system? Managing, we've heard that theme all day today, Dave, from Amar Al-Adha, all the way across. So hybrid clouds and hybrid management means one management system that sits on top of traditional, private, and public and the ability to say, I want to deploy my service and I want to say one definition of a service and deploy it in the public cloud or that same service deployed in the private cloud and I don't want to have to do things differently and that was what was really exciting at Discover. We unveiled the capability for, for instance, for cloud system to automatically reach out to a service provider and grab resources. In that case, we're showing that off with Savas. Your public cloud, your private cloud, deploying resources, but when you run out of resources, automatically starts bringing in resources from Savas. On an automated basis. On an automated basis. Completely automated. But Nick, so cloud management is a real differentiator for a lot of the cloud service providers and they got a lot of homegrown tools. Are you putting forth a scenario where essentially that becomes relatively standardized across the industry? For instance, we talked to the Savas API. So they still, each of these service providers still has their claim to fame. I know service providers that are building public clouds that are military grade, HIPAA compliant public clouds. Each one has their own value proposition, speed of which they deploy, types of places they deploy, footprints, compliance requires, I know I was working with one company and they need to deploy a service into Canada because there's a compliance regulation that that data must be in Canada. And so the nice thing about where we're going with this and having an overall management system, provisioning system that rides across it all is, I need to deploy a service in Canada, I do it. I need to deploy it in Singapore, I do it. Or if I've run out of local resources, it automatically gets resources from the public cloud as well. But your scenario is that you're able to provide that base infrastructure and then they can add value on top of it. Right, and our objective is to make sure that we show through the value of our partner service providers. So their value and their value proposition shows through so that when I set up a policy for this service, I can tag it saying I want it to be deployed into a military grade environment. And that means that one service provider that is military grade that is valid for this service to go to, but it can't go to this other service provider that doesn't have a military grade or HIPAA or PCI or such. So all of this ecosystem is really going to evolve and you want to be able to show that value from that service provider through the management system back to the customer so that they can pick and choose and have the right policies and bring in the right resources at the right time. Does that make sense? Okay, Nick, Van Der Spree from HP, Inside the Cube talking about tech, trends, cloud, cloud bursting, and really where the networking's changing. And I guess my final question to you is, will there ever be a hardening of the network? I mean, networking has come a long way, this conversion networking, we all hear about that. And Mike Bannick talked about it as well, but it's constantly under pressure. You know, smaller, faster, cheaper, but you've got more complex ways in the virtual machines that does that. So will it ever be hardened? Well, it's a good question. I'm probably, Mike's probably a better person to ask that level of detail to in his field. He's really been delving into that space. I just think that the door is opening up for us more broadly, I'm talking to more and more customers that say, I do have applications that fit into this world that they can leave my premises and go. And then the network bandwidth is continuing to boost and build the security. I know when we do bursting, you want to make sure that you're getting a secure VPN connections when you're talking to those service providers. Everything improves every day. And I know we're fueling a lot of that innovation in the industry today with a major focus on networking because that's the fabric that glues everything together. Okay, solve your sprawl, get to the cloud, management's key, thanks so much for coming back on theCUBE, Nick Manderspeed from HP, great story you guys have that ESSN division smoking hot, networking servers, storage. Thanks so much. You bet.