 Now, SiliconANGLE TV and Wikibon.org present a focus spotlight. Live from Las Vegas at VMworld 2011, host John Furrier and Dave Vellante, illuminating VMware integration with support from EMC, where cloud meets big data. Okay, we're back at VMworld 2011. I'm John Furrier with SiliconANGLE.com. We're here for blanket to blanket wall-to-wall coverage, Dave, with my co-host. I'm Dave Vellante from Wikibon.org and welcome to the spotlight on VMware integration. As you know, these spotlight segments are deep dives. We're bringing in practitioners, executives, customers, bloggers, and we're having in-depth discussions around particular topics. This one is focused on VMware integration. These are sponsored segments. This one is sponsored by EMC, and we're here with Chad Sackets. Chad, welcome to theCUBE. Dave, John, it's great to be here, man. You're a CUBE alumni many times, so great to have you back. Good job. Yeah, so, well, first question is, what face-melting action have you seen here at the event? And I know there's a lot of it. Well, so first things first, I'll tell you one thing I just melted my face. That was a great segment with Rick, man. The great customer story, you know, 300% improvement on SAP on VMware, on EMC using fast. My face melted right off, nobody. Face melting on theCUBE, Dave, that's a first for Chad. The things that I've seen, at least, that blow my mind, we've set new records around performance and scale. Loads and loads of customers that are now virtualizing the things that are heavy, that I'm encountering more and more folks where they say I've got confidence. They're virtualizing SAP landscapes, Oracle, those sorts of things. Obviously, in Steve Herrod's keynote, man, Applast, that blew my mind, right? I mean, I knew a little bit in advance, but that was one. And then, you know, I think one thing in Pat's, we ran a VM on an Isilon cluster. We moved VMs to run on PCIe-based Flash that had VMware integration on that in Project Lightning, previewing stuff to come, right? We showed something that's at today, we showed how an Isilon cluster scales out with changes in vSphere 5. I mean, just amazing stuff. There's one other thing that has just melted my face, which is we held a bunch of sessions, which are deep dives, like, you know, extreme nerd-outs where we're talking to the practitioner-level folks around do's and don'ts of disaster avoidance and disaster recovery and how they're blending and munging. You'd think like, how many people want to go to that? Well, at VMworld, I'll tell you, 600 people want to go to it one day and then it gets repeated the next day. There's another 700 people, so just awesome, awesome stuff. Yeah, it's been an amazing event, you know, back here in Vegas, we're going back to Moscone next year. We had Rick Jackson on, he tells us we're going back. You know why we're here this year? Because the parties are always really good here? Yeah, well, that's part of it, but I guess they're outgrowing Vegas is what I hear. So I understand from Rick that they've bought out the Moscone for the next 10 years, so. Yeah, it, you know, with the number of people here, even with the hurricane, right? It's very impressive, and you know that beyond all the nerdy stuff and all the fun, and you can see lots and lots of people doing fun things out here, and the Exhibitor's Hall is also awesome, but a lot of fun parties, like last night we had our customer appreciation party, gave out the journey awards and all that stuff. We had 150 more people than we were budgeted for or had space for. It was quite epic. Yeah, so we're talking about VMware integration. Chad, my first VMworld was only a few years ago, and I came in and storage guy, right? You start asking practitioners storage questions, and you find out right away there's a real problem. Oh yeah. And then you talk to VMware about it, and I remember they was saying, hey, we're doing our best, but we're resource constrained, and we got these APIs, and we're putting them out there, and we said, wow, this is gonna take some time. Time's flown by, hasn't it? You guys have put a lot of effort. Certainly the big storage companies, you guys, NetApp, IBM, HP, all working hard. You know, we've done the studies, and we've done the research, and we've published some independent material on this, and we went out and asked our customers, who's your number one VMware storage provider? EMC came up. Who's the best EMC came up? Not only within EMC's customer base, but across the customer base. Why is it that you guys were, why'd you come out number one? So first of all, it's a very difficult thing to study. It's a very difficult- Yeah, it was hard. You know, because first of all, there's a long list of things that you can integrate with, and second of all, that there's a subjective value of each integration. The fact that you guys did that survey, and as much as I might've tried to influence it, I couldn't. Well, yeah, I wanted to say. I mean, we came up with a methodology on our own, we did it on our own, you guys, we're not. And we brought in all the vendors, we studied everybody. We sponsored you guys. We sponsored you guys, but boy, oh boy, that was as independent as it gets. I would've tried, but I couldn't. Thank you for saying that, I appreciate it. It's important to say it. So basically, it's been a huge effort, a huge mammoth effort, and we're still, we've made lots of progress. So as an example, two, three years ago, four years ago, man, the problem of just basic storage scaling, people would run into very difficult problems around how many VMs do I put in a data store, and stuff like that, and they would create all sorts of difficult management challenges. So that required big changes, and that started to show up in VAI, which showed up in vSphere 4.1. Boat loads of support for that out of the gate, and now with vSphere 5, there's more. And so EMC early on supported that across everything that we do, right? That was a mammoth engineering effort. There's more to come on that front too. But the other big challenges around provisioning and integration around management and visibility. So when people would have problems, they wouldn't know what lives on what, right? And the VMware person would speak one language, VMs, data stores, and devices, and the storage person speaks in LUNs, files, systems, pools, whatever. So we made the first version of our plugin was two years ago, version one, of the virtual storage integrator. We're now on version five, so you can see like the rate of development and innovation is through the roof. Feedback from customers around it is great, but other innovations are coming online that are also really important, because when I asked the room of people who came to one of my sessions, how many of you use the plugins? Not only from us, but from other vendors. I'm always disturbed by how few it is, right? We showed data earlier on, and the brown here is the number who has it zero, and so you can see there's a long way to go here. It looks like a VU meter in this area. So the work around the policy-driven storage stuff, which if you talk to people around here, we actually got a panel where we were asking the room and asking experts, what are the features you're happiest about in vSphere 5? Storage DRS and policy-driven storage came up as like number one out of a whole bunch of them, because that policy-driven provisioning model is becoming baked in in the storage model. And that means that plugins are good, they extend those capabilities, but some of the stuff that we did in Gen 1 and Gen 2 and Gen 3 is now baked in, right? But again, the storage vendors have to step up, and that's a massive engineering commit. Chad, what are you seeing in the integration projects? Any new roles, because we're hearing a lot from people like in the big data world, for example, the role of assist admin, we got to make it easier. What assist admins are doing, it's been complex. All the integration projects are very different. You mentioned the diversity of projects. In the sessions and in your prep and all your work, are you seeing a new kind of role, not like assist admin, but this management's a big top issue, what are you seeing that you can share with the folks out there? That's a trend that someone might not be seeing right now. So what I'm seeing is that the infrastructure roles are getting munched together. Sometimes these, you know, we've been actually trying to help people with that with the whole cloud architect certification program that we've done. There's now 200 of them out there, 1,000 that are going through the program now. It's very popular. And what it's kind of a reflection of is that we need more people who are simultaneously deep experts, but also generalists. Because what happens is like in these storage models, you know, the VMware admin now has much more delegated responsibility over storage. Now, we're making it more invisible and more transparent through all these integration points. There's 75 plus and the list grows every day. But that person needs to know a little bit about storage and a little bit about networking, a little bit about virtualization, and they need to now start learning about VCloud Director and how to create infrastructure as a service. Very often that's one person or a small team of people that are built around someone who's a bit of a visionary. Yeah, I gotta ask you, we had the CEO at Tintree on and you and I have talked about this in the past, but I want to get your statement on record here. His contention was that we, we being Tintree design storage from scratch focused on, you know, the problem. That the big guys, you, NetApp, HP, IBM, the others are bolting on these integrations and as a result, you're at a disadvantage. What do you say to that? So, a lot of respect for Kieran, obviously. Smart man. Smart guy, yeah. Very smart man. Respect for startups in general. I came to EMC from a startup. I, in fact, it was startup number three. The first two blew up and then the third one EMC acquired. Startups need a few things to be true to succeed in my experience. In storage land, it's even more true because you're talking about persistent data, which is very difficult to get customers to move. You need to solve a pain point that is extremely profound. You've got to do it in a way where you've got enough sustained intellectual property advantage that's unique and difficult to replicate that you have enough window to succeed and you have to have the giants asleep at the wheel, right? Those are the ingredients that are prerequisites for any startup to succeed. Tintries, one, this is the year of storage startups around VMware. Nutanix, you know, others, all sorts of cool ones, right? And I wish them the best of luck but I have a news flash for them. The giants are not asleep. So... They'll either buy you or put you in the business. So look, they make an NFS server, an NFS server that integrates with vCenter. Hello, NetApp does that, EMC does that. We have vCenter plugins that enable provisioning, performance reporting, direct VM level object awareness. So the thing I'd say is since he put it out there first, I would have never gone here. You know, this is not polite and I'm a Canadian. So I wanna be polite. You gotta check him in the boards a little bit right now. I'm gonna check him in the boards here. In 2006, when he left VMware, his statement was totally true. People were struggling with, you know, a LUN that was 500 gigs in size and you could only put 10 VMs in it due to locking and management was terrible and there were no such thing as plugins or any vCenter integration, no VAI, no nothing. That need existed furiously in 2006. Now if you wanna have a NAS device that's VMware integrated, supports Flash, does DDoop, has VMware integration from the top down, 2010 called, NetApp and EMC do that and if you wanna have one that integrates from the top down, vCenter plugins and from the bottom up via native vCenter integration, hey, 2010 called, EMC does that. So your contention is the premise is the window is closing or has closed? No, the window's always open for startup innovation but the point there is that I think the core premise is flawed, number one that the problem statement of 2006 is no longer the problem statement of 2011 and the second thing is that the giants are not asleep at the wheel. Yeah, that's good insight. I talked to one of your customers last night, KGB, very memorable name and they were telling me about this capacity pooling thing that they saw which basically says they're gonna eliminate loans, make storage invisible like you've been talking about for a long time. I don't understand what people don't get. When I do the super sessions and stuff at VMworld, if you read between the lines, that's all real. So like for example, the other one is server compute mashups. I don't know if Pat talked about this, right? Well, a little bit. We're demonstrating running VMs on Icelon clusters and we've talked about it doing the same thing on Vmaxes. An Icelon cluster is 144 node, Westmere-based. So is there a room for server storage mashups and fluidity of that sort of? Well, of course there is. Ditto with PCIe-based flash with VMware integration which I'm sure the folks over there would say the giants are asleep at the wheel we're not asleep at the wheel. The other thing that I would say is getting back on a little more positive because I don't want to be negative, like I said, a lot of respect for competitors of any size. You gotta refute it. But I have to refute it, right? And it keeps you sharp, right? That's session that you were talking about, right? So I hinted about it in blog posts kind of going, it's got a weird name, VM application granular storage. It's scheduled intentionally at a really bad time, 11 AM on Monday, but trust me, go. And it was a technology preview of what we're doing going forward. And worked hand in hand furiously with the VMware team on that, anyone who went to the session would have seen that because almost, they demonstrated it on EMC, on NetApp, on Dell Equalogic, on IBM. But anyone who was there and watched the sessions, the first demo was in Unisphere, all the UIs were in Unisphere, everything, we showed all block NAS use cases. That's an engineering commit and comes back to this integration thing. The big idea of that is that as much as we've made progress, storage models for VMware are still not right. Even block NAS, that's the wrong debate. Ideally, storage devices would speak VMs natively. We, Tintree, are right now actually, the only two examples that I know of, where the array links to vCenter APIs, but it still speaks in files, or in our case in the language of files or blocks. The idea of VM volumes that Satyam and Vijay and Vinay showed in that session, to me that's one of the biggest things. That's a container with metadata, it's like an old mainframe volume table of content. So we have one last question each, and my last question for you is, last year they laid out the vision. This year it's a lot of new products coming in, demos, mashups, next year we're talking about, we're going to see more case studies, best practices. What are you going to see? So, I think the main thing that I think we're going to see is more and more examples of customers really using hybrid cloud deployments. That would be the first one. There's the beginnings, the whole global cloud alliance model was a big announcement, that the Colt, Blue Lock, Sonda and Singtel are kind of linking up. The other thing that I'd say is, I had a session around cloud application platforms. So whereas I get 1500 people showing up for disaster avoidance and disaster recovery, deep dive and storage and backup gets 800 people in it. Only 60 people registered for this session around cloud application platforms. Now stop and listen to this. This is the story of a customer who spends $400 million a year on cobalt programmers. They spend almost a billion a year on MIPS. The most transformational thing for them isn't backup. It's not DR. It's not disaster avoidance. It's about the application, right? So the project with them to use Gemfire, to connect it to their mainframe, is blows everything out of the water but only gets 60 people to attend the session, right? So I think people are going to year by year get it. That it's not just about cloud infrastructure which is important, will continue to be important. We're going to continue to innovate, out develop, out integrate all of those things. But the application layer, like the vFabric data director thing, that's huge, huge. You know, I think we'll see more and more people kind of understanding how important transformational that stuff is. All right Chad, well we are one of the few people that could go longer with us than Pat. But we're out of time, unfortunately. Thank you so much for coming on the queue. Dave, it's my pleasure. Really appreciate it. John, thank you. Good luck with everything. Thanks Chad, Chad Saki with EMC. Okay, and we will be right back with more coverage from VMworld Live.