 Okay, we're back live here at HP Discover Day 2. This is theCUBE, our flagship program. We go out to the event and extract the signal from the noise. HP Discover in Germany, in Europe, record attendance, growth, HP's got a great event here. This is SiliconANGLE's exclusive two-day coverage. I'm John Furrier, the founder of SiliconANGLE. I'm joined with my co-host. Hi everybody, I'm Dave Vellante of Wikibon.org and we've got many-time guest, Craig Nunes, who's the worldwide VP of marketing for HP Storage. Craig, welcome back to theCUBE. Thanks. Always good to see you. He's a cubie. He's a cubie. A cubie is a cube alumni, so great comment, great idea there. A groupie of the cube. The cubie. Yes. So, big day on Monday, we flew in, we saw the press conference packed house at the press conference. It was a good crowd there. I hosted the customer panel, I had five customers, and we're going to talk about that. But the announcement's big, Craig. I mean, you guys have been working real hard to rationalize the portfolio, communicate what you're trying to do, get new products out to the market. Bombs are dropping and you're fighting the competition. This really was a big coming out party for you. Speaking to the folks on your team, everybody's really excited, and a lot of work went into getting to where you are now. So describe where you are now in your view and then we'll talk a little bit more about the announcement. Yeah, so, I mean, we've talked about this in the past. We've been on a journey, right? For a while now, and it's something that started really several years ago, and I think probably the coming out party, mid-11, when we kind of introduced the idea of converged storage, storage built on HP's own IP, really aimed at kind of the biggest areas of growth for our customers, try to handle that. But what was significant about Monday was really the other shoe dropping. It's all of the investment we were packing in and the balance of the strategy that, to be honest, was more meaningfully told with these new product announcements, we were able to bring those forward. The bottom line is that announcement was all about simplicity. And in fact, we talk about something called polymorphic simplicity, which is many simplicity that really is available in many shapes and sizes across your storage infrastructure to solve the problems that our customers are dealing with. And today, believe this or not, because in preparing for what we're going to talk about, we did our homework and our customers deal with upwards of 10 architectures just to get the storage job done for their business, 10. And who is single source these days, right? So 15, 20 architectures to get storage done and bottom line, big cloud, sorry, big data, cloud, all that's going on in IT mobility. Our customers are saying, look, stop, I can't cope with that and fight the battle with 15 or 20 things that have to change through this, right? So we're looking for some difference. So poly, many, morphic shapes and sizes, simplicity, we all know what that means. So polymorphic storage, let's face it, you guys are going after EMC's underbelly when you do this. And EMC's got a lot of different architectures. You put up a slide that said, okay, we've got all these, you know, VNX and VMAX, et cetera. I think they had, I don't know, you had seven or eight in the slide and you could probably count up a few more. We had 10 in the slide. You had 10 in the slide? Okay, so you're going after that and saying, okay, we have an alternative. What is that alternative? Yeah, so the bottom line is for primary storage, one architecture, low to high across file block object, across traditional hard disk drives and solid state, one common set of services, one platform to get the job done. And that's for primary. And for information retention and information protection, one architecture, common, low to high, common services, et cetera. Now, in that realm, we believe information retention, information protection are coming together. Most still buy is separate kind of motion. So we have separate products that we offer today, but they are built on the same architecture. The same duplication that runs in information protection is going to be the same duplication that runs in information retention. It will be the same services. And by the way, the same duplication that runs in primary storage, right? That's the whole deal. So the bottom line is for our customers, it's go from 10 to two or three with common services, low to high, one management platform that's going to handle all of what your storage needs. And when it comes to new things like solid state and flash, HP is not a company who's going to run out to buy something. We've got the platform to deliver it. We're not going to add to our customer's pain in introducing capability like that. So less is more in this paradigm. In a big way. So, okay, but at the same time, there's no God box. You can't have the same box to primary and backup and archive. You've got to have different platforms that. So there's still a horses for courses requirement in the marketplace. We can agree on that. Yes. Your strategy, let's start with primary, is a single architecture for primary, which is really built on three par. Built on three par, the success we're having there, yeah. And you're attacking your competition, EMC in particular, but there are others out there that are putting forth this pure horses, I say pure, more horses for courses strategy. We've got this product for that problem, this product for that problem. Three par is your answer for primary, right? That's what's now called store serve, right? Three par store serve. Now, people say, I would say, well, wait a minute, still got a bunch of other architectures. Got VBA, you got MSA, you got what you call a store virtual, which is left hand. You have all these other architectures, isn't that still quite central? Yeah, so, yeah, so great question. Here's kind of how we look at it, how our customers look at it. We've got customers where they are today, right? Where HP was a couple of years ago, right? Got a lot of VBA out there, buying XP today. And those customers are in various states of readiness to get to what's next, to get to converge storage, right? And we're not going to leave those guys high and dry. They've got needs today. They've got a business to run, service levels to maintain. And so for those guys, we have what we call our established platforms, but for customers who are ready to move, or frankly, for the market we're not serving today, and it's a fair bit of it, that HP is not the storage choice. It is converged storage. It's three-part store-serve. It is store-all for information retention and analytics. It is store-once for deduplication and backup. You mentioned store-virtual. That's another great segue because part of the approach here is not just delivering for storage systems, but we have a set of customers, a set of CTOs who are thinking a little differently about their infrastructure. They will only take on storage services in a way that's completely decoupled from the underlying hardware infrastructure. It's a cost issue. It's a lock-in issue for them. So, a more software-led approach, software-defined data center, software-led data center approach is where they're headed. And our lead platform there is store-virtual. Store-virtual, frankly, was born as that before the term was coined. It's been in the market as a virtual storage appliance. That's how software-defined storage will be delivered. And the idea with software-defined storage is obviously it's got to be software-led. It's got to be able to spin up a controller in a VM or lay it down on a standard X86 server. It needs to, so we'd like it to run on our servers, but it'll run on Dell, IBM, whatever, our current store-virtual VSA. It'll run in any hypervisor, right? Got customers who are, got a couple of hypervisors running. Again, the lock-in factor. So, taking an open approach there. And what we're finding is, call it, the hardware is caught up. You know, why now? We've been doing this for five years. Processor, horsepower, huge, right? There are cycles in the server that aren't getting used, that can be tapped for this. Drives, three terabyte drives, times however many in a server. There is capacity in the server wasted. Spin up software-defined storage, take advantage of what's on the floor. We've seen people save 50 to 60% of their power, cooling, floor space by deploying this stuff. 70 to 80% of their storage costs. So remember, they're basically tapping something they've already bought. And for these guys, a very easy way to grow, right? It is polymorphic simplicity, right? Software, systems. So Craig, polymorphic. So this came up yesterday, the whole hypervisor thing, Ignostic was a big message. We had Sam Johnson on, he's a friend of the cube, you met him. He's kind of an analyst, he's a blogger, but he's kind of an analyst. He works at a big company, but he's really well regarded in the cloud space. This hypervisor-agnostic thing is a huge deal in the cloud space. Now you guys are talking about that storage. So that's very, very cool. The one thing that I brought about on my blog two days ago in the announcement that I found really compelling was that you guys have lightning in a bottle. I mean, this whole software-defined storage direction is really a great pivot with software-defined networking because that's where all the buzz is. But you guys have actually three sets of products there now and so that's really, really relevant. So I think you guys have lightning in a bottle with the positioning with storage, networking, and servers. The other thing that we've talked about with autonomy is in the big data space, there's a big conversation around where to put the data. And so what you guys are doing that's innovative against the other guys and your competitors is that you're putting the data near the storage. You guys are thinking about the data. So I want to ask you, how do you look at that and what's the current plan and obviously the new horsepower, the new 7,000 is a whole nother level of product at a price point and all those good savings, but relative to the big data explosion. I mean, autonomy was up on stage with you guys. You didn't really talk about that because it wasn't part of that big announcement, but it's such a relevant piece to your piece. Can you share the big data story and how that fits with software defined or software led? Yeah, so to kind of catch everybody up, the announcement was about bringing a no SQL database inside of the storage platform, inside of your unstructured, your storage of your unstructured data and why so that you can get at your hundreds of millions of files and if you start going through it, people are closer to that than they might realize. To search that, and I'll give you an example, if you've got 500 million files and if they're 100K files, guess what, that's only 50 terabytes, right? We're not talking about a lot of capacity. 500 million files will take you 42 hours to find something, right? You got somebody consuming a ton of capacity to the point where they're burning through your buffer. You could well be at a capacity before you find out who is doing it, right? If it takes you 42 hours to find it. Express Query, 1.4 seconds on 500 million files, right? So that is the only way folks are able to stay ahead of the massive unstructured growth coming in or taking advantage of data that's not their own, right? You know, I'm a marketing guy. I take a look at Twitter feeds, social media feeds and the only way to keep on top of that is to be able to mine that. Instantly. You know, David Belonte and David Floyd will go into the weeds and talk about all the speeds and feeds of the storage, which is cool. I let them do that because I don't want to but they do it. I love the technology. But what I'm impressed by is the holistic view that you guys take with the market. Like it's very rare to see a storage focused group look at the marketplace from a holistic perspective and saying, hey, we got to think about storage from the perspective of capacity and performance, okay, table stakes in the storage business, but hey, we got to have low latency analytics, support the analytics, support the big data piece because that is where the growth is right now. So one, is that, what is your holistic view there? I mean, do you look at big data as, hey, do we just got to store some stuff or what are you guys doing in the software particular that speeds that up? Yeah, so the approach is kind of product management 101. I mean, I don't know that I'd call it anything crazy, but it's just kind of tuning in to the pain customers have with their unstructured data or content depots or whatever you want to call it. And the fact that, I mean, we have customers today who are putting up with loads of files the workflow it takes to get through it. We are a little different though. We are able to go partner with our brothers at HP Labs and sisters and when we kind of talked through the problem with them, they were the ones who kind of helped us with this notion of bring this tool into the storage platform and what we learned in working with our autonomy team is what they go through to find and filter data, it'll take 10 servers four days to do what we're doing now with one server in a few seconds. So suddenly, number one, it's kind of, hey, convergence. We've been talking about it, it's real, it's happening. It happens to be all happening on store-off. So other vendors sell a bunch of platforms, right? So you guys said, okay, you're one platform, but you actually have a lot of other platforms that you're supporting, but you're driving all your new customers to this new platform. So it's not like you're selling, you're still supporting and selling, I guess, upgrades. You're not onboarding new customers to the old stuff. So you're exactly one platform out there. That's right. The three-par, poor tech. And can you explain that? Because that's hard for me to understand. So, and kind of to a point that Dave raised a minute ago, we've got one architecture, one thing for primary storage. Again, whether it's solid state or hard disk, whether it's object or block, it doesn't matter. We have one thing for information retention and information protection, but the services that our customers will touch, the data services, those are going to flow across both, right? When we're doing encryption, when we're doing deduplication compression, same approach across both. Management, common across both. But the underlying architecture, we've got three-par and we've got the store one, store all stuff. And that is, we think, the way of the future and we've got it today. The vision David talked about, we deliver right now. So, it is vision, but it is real. It's, you know, what's the big announcement with all of that. It's a shipping product. Okay, but I need you to talk about SSD Flash, because you basically, you have this primary storage architecture. Flash is changing everything in primary storage. Are you going to just dump Flash on top of three-par and three-par handle that? Is that the Flash platform? What is the Flash platform? Yeah, so the Flash platform unequivocally is three-par. Why, how? Look, the reason HP picked up three-par a couple of years ago was because it is a modern architecture built for all of the stuff that's going on out there. And if you think about what is going on in Flash, there's a lot of small companies, right? I love that. I came from one of those, right? But I've seen that movie and let me tell you how it ends. It can end three ways. First, you go out with a killer feature and it's probably not Flash, because guess what, we all buy from the same guys. Yeah, it's not the Flash itself, it's not the media. Yeah, but you have a killer feature and the incumbent, the big guy says, interesting, I'll have it in a year and they freeze the customer. So you die by inertia, right? Number two, it's about software, not the Flash. So what are the data services? Where's the QOS? Where's the DR? Where's the HA? I mean, we're talking about half a million IOPS on a box. You better be protecting that or you're losing who knows how many applications. So where is that? It's not an investment issue, because as a small company, you're well out-invested by the big guys. It is probably an architecture issue. So a lot of those guys are in the market to work out how they solve a problem for a big guy. Extreme IO, EMC, perfect example, by the way adds to that complexity picture we painted, right? That was number 10. That was number 10, the most recent. And so if you are in the market without those data services, hot box, I don't know what kind of traction you're going to get. And I know one of the guys out there builds themselves on able to deliver quality of service, pin it to the application. Guess what? We introduced that Monday, right? Not just pin it to the application, we'll pin it to the tenant and we'll support multiple tenants' applications on the box. You can guarantee QOS or you can over provision. The three-part is the future of the solid-state platform. It is the future of the solid-state platform. We introduced a solid-state version of the 7000. We also announced that a optimized solid-state platform or flash platform is coming on the three-part architecture. All right, Craig, this is always great to see you. Appreciate you coming on. Congratulations for the announcement. Thank you. Good luck with you watching. Okay, this is theCUBE. We are here at HB Discover. Big announcement from the storage group from Craig and his team. A lot of press and congratulations again and we'll be right back with our next guest after this short break.