 Okay, we're back live at HP Discover 2012. This is the Cube, our flagship telcozoo out to the events. I'm joined by co-host Dave Vellante. In this segment we're going to do some analysis on HP's position in the converged infrastructure market. Dave, your team has done some real amazing work around converged infrastructure. I would keep on .org, a lot of free content there, but mostly what's really kind of telling is you guys estimate the total available market close to a half a trillion dollars. So, converged infrastructure is a huge market, but that's networking, storage, and servers, and that really is what cloud is. And this is our huge market. So, what's the analysis on HP? I have my opinion, but I want to hear from you first around. And I would add the underlying software and services. And so, first of all, HP's shoring up, it's two businesses in terms of networking and storage. It's really done a good job there. It's always been strong in service. It's got some very cool stuff going on in storage, but it's not done, right? So, it's got to prove that the store wants to do great marketing. It's going after EMC hard. I like that. And very not HP-like, used to work there, very conservative, but so the marketing is going right at EMC. It's causing some angst to EMC, and so I love to see that. So, the good discourse going there. HP's got to show that it can gain market share in that space. The second is flash, right? That is going to be a hugely disruptive space. So, just like three-par disrupted the tier one space, flash is now coming in and is really trying to become that new tier one. So, we're seeing a whole new change in the way that storage tiering is occurring. Yeah, and if you look at the market, Dave, one of the observations I have is that we were just talking to the folks at IBM. They were once this little storage, little siloed organization. They've now recognized, as we've been talking about for two years now, since we quoted storage as sexy at EMC World 2010, storage is at the center of all the action in conversion networking and servers. And storage is really where the action is. As with big data, becomes this massive thermal growth in the marketplace around analytics and transforming business. Storage is really the core linchpin in the enabler. And it's moving from spinning disk industry, supplier, boxes, hardware to an absolute enablement. So, storage is the critical piece of this new architecture. And what's interesting is that you have legacy approaches and you have kind of new thinking. So, one thing about HP that's interesting is that they're changing the conversation around tiering and just looking at things differently. And they're delivering the value. And they're being aggressive, as you said. They're press release that on Monday, like a little nuclear bomb in the storage business. All these claims, and they're backing them up with facts. So, I mean, what's your take on that announcement? Well, so first, the big thing before I got an announcement is the big thing is HP has become relevant in storage. Three or four years ago, HP was not relevant in storage. They had to do that through acquisition. That's fine, I've said acquisition these days is oftentimes a better investment than R&D. In terms of that announcement, look, EMC's got two thirds of the market. Somebody's got to take on the market leader. Right, I mean, it's sort of unnatural. And who's got two thirds of a market in any storage business? It's just, you know, is it sustainable? That's the big question. So, I like to see what HP's doing. And really, HP kind of stands alone there. But I think you're going to see some changes there as well. I think you're going to see some new approaches to backup and dealing with backup windows, which are going to have further disruption. What I really like about the HP strategy, it's got a deduplication, and this is nuance, but it's got a deduplication algorithm that goes across the storage portfolio such that you don't have to rehydrate, meaning undedupe and rededupe to move stuff around the network. That is a great vision. And really, nobody is delivering on that. HP has definitely got the potential to deliver on it. What's EMC's answer to that? Because obviously, EMC is a different perspective. They're going to jump up and down and say, you know, wait, that's just BS. So, ironically, the first place I ever heard that vision was at EMC World prior to the data domain acquisition. Now, once EMC bought data domain, I said, forget that vision. That will never happen because you've got Avamar, you've got data domain, completely separate algorithms. They are trying to bring that together and glue it together with user interface. EMC is now going to try to change the game, in my opinion. They brought in Stephen Manley, who was the CTO of NetApp. Now NetApp's got a completely different vision around backup. The problem is NetApp doesn't own any software assets. They got a partner with guys like Syncsort and Commvault. And I've talked to Tom Georgins about this. He said, you got to buy some software assets, and he's reluctant to do that because he doesn't want to compete with his partners. Personally, I think that's a mistake. I think you've got to increase the software value added in this business. And what I'm talking about is the ability to use snapshots to take a continuous data protection approach, like an Apple time machine for the enterprise. That is the new paradigm for backup, which I think has a lot of potential to disrupt that business. Dave, one of the things I've been blogging about, and we've been talking about on theCUBE, is that with big data, all these architectures are changing in the network. Certainly the convergent infrastructure is obviously a big part of that equation. What does HP have to do right now? Obviously they bought autonomy. I'm interested to see how that affects some of the architectural issues around storage. What does HP have to do to really take their storage mojo, which is really strong leadership right now, and take it to the next level? Well, I think for years, John, the system vendors, particularly IBM and HP, have sort of been threatening that their system prowess is going to lead them to dominance, and it hasn't happened. Now, what's changed is converged infrastructure. Where you've got servers and storage and networking and end-to-end system performance. And the other thing is you're starting to see the flash hierarchy move from the array into the server. I think that favors the server trends. Now, to EMC's credit, they're not sitting idle. They see that's happening. So they're making moves. They're late, but they're trying to compete with Fusion I.O. They buy Extreme I.O. So that's going to be a really interesting battle. Do you see the business models changing in storage? Obviously I'm oversimplifying, but I see the work attached. Storage has always been this kind of, oh, it's attached to servers. And we were talking to Intel over the past couple years and past weeks about storage was kind of subservient to the server chipsets that people would buy in bulk and storage can never get the credit. But now, storage is actually with the growth of big data changing things. Do you see the business models changing with storage where it's much more integrated and or with flash, these new organizations? Can you elaborate on that? Yeah, the more things change, the more they stay the same. I think one of HP's biggest opportunities, John, is to do a better job of connecting to its own servers. It actually has a relative to IBM, for example, a lower share of its server attached. So if it just does that, it's going to gain share. Now my opinion is the focus on storage hasn't been there at HP historically. Now they bring in Donatelli, he brings in a bunch of his buddies from EMC, Dave Scott has shown that he's got continuity there. I think that continuity from an organizational standpoint allows HP to be more focused on something as simple as selling its own storage with its servers. Okay, so here's a question for you. I want you to just put the Dave Vellante prediction hat on. What do they do with autonomy? What does the storage group do with autonomy? Because autonomy is a business sell. It's not a real box and or typical HP-like sell. It's really more of reference architectures and it's a software company. So what do you think autonomy does to the storage group? Does it change at all? The biggest challenge that I think that HP has with the autonomy acquisition is autonomy is software and the core HP Salesforce doesn't really know how to sell application software to CIOs. However, I was impressed this week and we've talked about this, the number of different groups printing and storage, the server guys who are seemingly having serious integration with the autonomy asset. If in fact they can succeed in doing that, that's going to make it a lot more comfortable for the core HP Salesforce to sell autonomy. Here's my prediction. I think that, and this is just pure vision at this point, I haven't heard anything just as my opinion. I think that with the big data movement coming in with the software, we've said on theCUBE that software's the key to sitting on top of all these systems, right? So the software's obviously really important. I think there's going to be a breakthrough in storage because storage is so important that when you bring in that kind of autonomy software into a storage portfolio, I think the R&D guys can really probably do some pretty good work. So my prediction is that a breakthrough technology will emerge, some sort of combination of flash, new tier, whatever, something is going to come out of that. I feel like that could really be valuable. Well, and I think from again, from HP's standpoint, the big challenge that it has is organizational. I mean, for example, HP's data protector software is in this software group, they're trying to build up the software group. The key is can HP collaboratively work across those divisions, package that software like EMC does and start selling things in bunches like bananas. That is the key, hardware, software and services go together and that drives margins that drives cash flow. What do you think about, we've been hearing flash everywhere, flash this, flash that, what is the flash trend due to storage? Because when we were talking about flash, not being a price sensitive issue when you use it in a storage architecture. Obviously the cost of flash needs to go down for full ubiquity, but in storage, flash changes the economics. What is your take on HP's opportunity with flash and what do they do? I think flash generally is a huge productivity impact to organizations and that's why I'm not as concerned about flash pricing. Yes, the average day to day storage practitioner is concerned about flash pricing, but flash is going to have huge productivity impacts on organizations. They're going to pay through the nose for flash. Why? Because they're going to drive business. I think flash is going to completely change storage tiering, where tier zero is now the server side flash like Fusion IO for example and others and I think flash, all flash arrays become the new tier one. And I think the traditional tier one slash 1.5 completely changes. I think the dynamics of that business change and become more like what 3PAR has, multi-tenant cloud application support across the portfolio, not siloed and single application purpose built. And I think 3PAR is very well positioned there. My question is, is the Symetrix Vmax 40K the last mainframe class storage we'll ever see? Pat Gelsinger, we asked him that question. He said, oh no, we've got a lot more coming. Yeah, Brian Gallagher is just a cool guy. I really think that that class of storage is going to get driven down and be more competitive. I think what's happening is Vmax is trying to become more like 3PAR. And now the other thing is Extreme IO is a very interesting acquisition. Look where EMC put the Extreme IO asset. They didn't put it where logically it should go in the tier one Symetrix group. They put it in the Palatano's group. Why? In my opinion, because it'll be competitive. I think Gelsinger was concerned, if he put it with Gallagher, they wouldn't get the features, it wouldn't get the care and the feeding and they'd kill it. The Palatano will be competitive with Gallagher. And I love that kind of thinking. That's some good inside baseball there Dave. Great analysis. Final wrap up point I'd like to get your opinion on is the Converge infrastructure team at HP. They got the storage, they got the networking and they got the servers. We saw Gen 8, we've seen some good and great work with the storage group. The network seems to be the bottleneck although some people will disagree the Paulienist agrees that the networks are the bottlenecks right now. What is your watch list for HP right now in the Converge infrastructure? What are you looking for? Well as you know I said for probably a year or so that it was a two horse race between VCE and HP and that's changed. Now of course HP says we were first, VCE says we were first. I like what VBlock did. I think VBlock was very disruptive and innovative and I think HP got out in front of marketing and I think it's really begun to now get its act together. I think it's got some ways to go though in terms of that whole logical block, single block architecture. I think it's doing great marketing and I think its products are now catching up but I will tell you right now, the server side HP is great. As I said they've got some really cool stuff going on in storage and networking. Obviously we're going to have Bethany Meyer on and I think that piece is coming together as well. Yeah I think just to close here I think the Converge infrastructure group of HP is really poised to take it to the next level because the world is going to see a winner come from either the multi-vendor side or the absolute proprietary purpose-built side like Oracle. So it's Oracle versus the SAP HPs of the world so I think at the path that they're on right now they have a nice great roadmap into cloud and I think that multi-vendor with their partnerships and their channel HP's got a great opportunity with. And I think here's my prediction and within the next five years more than two thirds of the infrastructure that's sold will be of the converged infrastructure variety. Either as a reference architecture or as a single block SKU. Converge infrastructure at the heart of the solution storage is at the ground zero enabling it also that's great analysis. Move back with the cube here at HP Discover live in Las Vegas right after this break.