 Live from the FIA Barcelona Grand Via Compensator in Barcelona, Spain, it's The Cube at HP Discover Barcelona 2014 brought to you by headline sponsor HP. Here are your hosts, John Furrier and Dave Vellante. Here at Barcelona, Spain for HP Discover 2014, it's The Cube, our flagship program. We go out to the events, extract the sound from the noise, winding down day two of three days of wall-to-wall coverage. The Cube extracts the sound from the noise, connects the dots with the crowd with the crowd chat. I'm John Furrier with my co-host Dave Vellante. Our next guest is Cube alum, one of our favorite guests, Craig Nunez, who loves to come on and give us the update on HP Storage. He's the Vice President of HP Storage and also a good friend and supporter of The Cube. Really be thankful for your support over the years, appreciate it. Let's get down and dirty right away. Party last night was really good. We found the recipe in Barcelona. We found one that works. Sucked the oxygen out of all the other parties. Good time, good job. Good to see customers there, Brocade, Emulex. Yeah, I know it was great just to chill out with everybody. Storage is not just a product, it's a way of life, right? So we're just living it in Barcelona. You guys have a great community. I was just saying a day about commenting. Storage, you guys really invested in and built a great community, a three-part culture when you guys had it before you came to HP. And HP had some social head Calvin doing some stuff. They're doing some podcasts. But that was the beginning of what becomes known as community. It's community on the developer side. But the customer community, you guys have great loyalty, brand loyalty. Products are great too, but more importantly, you have customers down there dancing. It really shows if you guys have not just bought community or tried to check a box and some program called Build a Community, you actually have done it the right way. So congratulations. Yeah, thanks, appreciate it. What's the secret of success? Share your secret sauce. God, I tell you, if you can build stuff that delivers, that works as promised and make your customers heroes and then connect those guys with the engineers with the technical guys so that they're a part of what we do next. Everybody becomes emotionally involved in the success of what's going on. You guys are great. You do the cube with the executive level from the business side. You got Calvin and the bloggers in the blogger lounge. They have tech talks going on all around the clock. That is the secret. Just have a great product. Do what you say you're going to do and connect the right people to have the right questions and the right guys with the right answers. That's right. All right, so let's get into the products. Let's talk about competition. Where are you guys winning right now? Because David and I are talking about, you know, the reason to be the architecture you guys have versus some of the competitors. Some startups are worth $4 billion, like Pierce Storage. You got EMC getting in the flash business. You have Flash. What's going on with the competition? How do you guys stack? Well, an IDC just published a rear-view mirror if you saw that. You guys weren't first to market with the all-flash array. And now you're growing like this. Absolutely. So the numbers are going to change. Yeah. So we have seen kind of the hockey stick in Flash and the pole for it. And part of what we were talking about, in fact, we talked about it, I think, last June, if you could get the affordability right on Flash, if you could drive it down and, you know, crossover spinning discs, people will buy it because who wouldn't? It's faster and better service levels, right? So we had been, I won't say singularly focused on that, but that's been a huge, huge focus. We rolled three generations of discs, a ton of software investment, crossed over 15K last June. And the, you know, it's shown in our numbers. We've shipped a couple of quarters ago more Flash capacity than 15K capacity. You know, the next stop is crossing over 10K capacity and, you know, continuing to drive down the cost of Flash aggressively. And then, you know, the other side is simply delivering the scalability, many times the scalability of what the other guys are delivering out there. I had one customer here at the show say, you know, one of the factors in my assessment was I would have been out of capacity day one with the other guy. And, you know, with yours, you've got six times that, it was good for a long time. You know, so focus on scale, focus on the tier one availability, resilience capabilities, it's been good. And because of that, you know, there's been a huge amount of pull because people are coming off high-end storage systems where, you know, seeing a lot of VMAX customers who are kind of fork in the road. Do I go next VMAX? Do I try to, you know, take a different approach? And they love the Flash story. Why? Great service levels with Flash, all the tier one availability you need. And the space required is like nothing compared to the, you know, the tiles and tiles of spinning drives and VMAX friends. So around the time that you and David Scott left HP to go start 3-par, HP kind of turned off the R&D spigot in storage and said, ah, we got EVA and let's just bump along with that. And that opened the door for EMC and NetApp to come into your install base. And then of course the 3-par acquisitions are documented and all known. You should now be in a position to gain share just within your own base. Are you seeing your share of market for the 3-par piece of the business dramatically, you know, meaningfully increase relative to that period of time with the spigot shut off? And what would you expect for all Flash? I'll give you my perspective and you know, this is kind of Craig's view. And I was not at HP back then so outside looking at... Well, you weren't in storage either, right? Well, I was in storage while, so HP picked up Compact, bought a nice storage business. I was at 3-par at the time. And even after that acquisition and after some of the poor investment choices that were made back then, I think HP overall from the storage perspective, they continued to gain share. They were top in the market for a long time. And so the point to that is the backward looking share metrics don't represent what's happening today right now with the customers. It takes a while to catch up. And it took a while and once things kind of turned south, very clearly that things were in bad shape. And so I think the work involved in kind of retaking leadership, like here in EMEA, I love coming here because in the mid-range, you've got that leadership. 3-par is the number one mid-range. Well, you didn't have the hot products back then. You got the hot products. Yes, and so part of it is just good old-fashioned proven it out. Like we're talking with John, every quarter delivering something better, getting folks... I guess the reason for my question is that... Yeah, so the reason my question is to gain share, you don't even necessarily have to go whole hog off-platform. You could do it on-platform. Is that a correct answer? When you say on-platform, you mean attaching to HP... HP customers, yeah. Well, I look at it different because HP customers, HP server customers have made a storage choice at some point they were happy with. And we'll pick on our buddies at EMC. EMC's got a lot of good footprint out there. And so I don't look at it as an HP server attached. I look at it as an EMC displacement. And so we've got to bring markedly better stuff to get folks to consider something different. Now, the good news is with the transition in the mid-range that EMC's been in the process of going through and now the transition high-end, there are points where customers are trying to work out. What do I do next? And the easier you can make it for them to consider something else, the better. And we've been focused not just on great 3-par capabilities, but solving the problem of moving data from the EMC platform to 3-par. In fact, here at Discover, we introduced a data migration capability built on our storage federation software called Online Import from VMAX. And it just makes it easier for folks, lowers the risk for folks to come on over and take advantage of it. Well, Lee Petlaw was on. He said they did a bake-off at Sony. And he said they looked at performance, scalability, resiliency in TCO. Those are the four metrics. He said 3-par all-flash array. Go it away. And in particular, the performance, I asked him, could you see a meaningful delta in performance? He says, oh, yeah. A lot of the all-flash arrays, when they filled up and they had a good garbage collection, performance went to hell on invested. Why is that not the case for you guys? Well, part of it is they are going to, it's back to kind of a scale question. You're going to hit your 30 or 40 terabyte raw max in a lot of these platforms. And we can keep on going to 460 terabytes raw. 1.4 petabytes after dedupe. So you're not going to hit those scalability limits, those algorithmic limits in your dedupe algorithm and memory, you know, period. I mean, you're going to have a platform that's going to scale. And we've never suffered from, even with spinners, we've never suffered from as you fill the platform, you know, your performance starts to degrade. We don't have that issue with 3-par. We have a fundamentally different architecture. And, you know, a guy like Cnet in his areas, a guy who could explain to you in no uncertain terms exactly how we get that, but we simply don't have that issue. So, you know, that for sure is a part of it. And it's, you know, it is, it's not one thing, but it's, you know, the performance discussion, it's the scalability, it's, hey, I'm putting my most important stuff on this platform, you know, what are you doing about single system resilience and availability and all that stuff, and it's there. And then, you know, and then when it comes time to actually, you know, go through a transaction, the cost structure is great. I mean, we have a cost structure that no one can beat today in the industry. We are out ahead of folks with the large capacity drive, and, you know, we're going to keep hammering that away. Well, and you've got, you're the most robust all-flash stack in the business. Right on. I think that's a, I think that's a, I guess that's my opinion. I'm trying to think of anybody else that stuff is new, so they haven't had the time to build the stacks. IBM with TMS, connected to an SVC, it's got a nice stack, but you got to have an SVC. If you don't have an SVC, it doesn't have a stack. So, that to me is a big advantage. I want you to talk about the persona thing. Sure. You turn a three-par into the God Box. God Box. I hate that word only because box is in the word, and it's there. Nothing to do with the machine. The God solution? Yeah. It's all about the software approach. And the cool thing with personas is we have evolved the software to take advantage of a file system and rich protocols on top of that and rest-based APIs right in the OS. So, it is the same thing that served up your VMware environments and KBM environments and your databases, et cetera. The same software is now serving up you know, SMB and NFS and rest-based data. So, it's the same architecture, the same platform. And so, manageability is way better. Folks love what they go through. Efficiency is better. What I have been telling folks is what personas is not, is it is not an emulation layer. A file emulation over block or block emulation over file is one of the vendors who has implemented it. It is not a management UI over totally different architectures. It's not that. And that doesn't really help. It kills you on efficiency. It kills you on density. It kills you on management time because you still have to do work. And so, we have the benefit. I mean, let me also say the trade-offs the other guys made, who cares? They made a great unified storage market and folks were willing to deal with some of those trade-offs because they liked what they were getting. I think what we're bringing is hey, you know what? You don't have to deal with those trade-offs anymore. You've got a great unified storage platform. You don't have to take a look, right? Well, the conventional wisdom always says it's easier to go from small to big, Intel, right? You can't go from, you know, large mainframe class down. 3 Pro has actually proven that wrong. You know, you're really doing it. The other thing I'll observe, we're getting the hook. There wasn't a brain drain after HP acquired 3 Power. You kept all the really, really key people. Well, if you think that we have a cult like following in our customers, you know, talk to the engineers, talk to my product managers. You know, those guys, they live and breathe this stuff. They bleed this stuff. It is, you know, it's more than just a job and a product, right? This is like, this is them. And so that has really, you know, held together over time. And the way, you know, the way we run the business and the part of HP is, we get all the benefit of the HP umbrella, but we continue to drive, you know, a great focused storage business. You know, around 3-PAR, Explore Virtual, Explore Ones, and the rest. So it's been great. All right, cool. We're getting the hook. I'll give you the last word. I really appreciate you coming by. You know, it's just super busy. We're packed. But we can't, it wouldn't be the same crap. So I'll give you the last word on Discover Barcelona. The storage space here. What's your bottom line? Yeah, so we've spent the time talking about Flash, but we are getting a ton of traffic into other areas. The software-defined storage area, because it's not just about VSA and VM-based storage. We've got a wonderful, hyper-converged appliance built on that same technology that is hot, hot, hot. And then we're seeing a ton of interest in what we call flat backup. This approach to really turn and store once into application-managed backup feature of 3-PAR. For the, at home, my family are Apple users and they have Time Machine on Apple. And what we've got here is a lot like the Time Machine feature for 3-PAR. And that has just got people's eyes wide open to understand how more easily they protect their 3-PAR data. That's driving a ton of... Well, that model, that paradigm, I've been talking about this so I can't tell you how many times. I'm really thrilled to see that happening. I'm not surprised that the uptake and the interest is very high. I mean, it's the right way. All right, thanks again for coming on. Always a pleasure to see you. All right, keep right there, everybody. We'll be right back after this. This is The Cube. We're live from HP Barcelona. HP Discover from Barcelona. We're right back.