 From London, England, extracting the signal from the noise. It's theCUBE, Covered, Discover 2015. Brought to you by Hewlett Packard Enterprise. Now your hosts, John Furrier and Dave Vellante. Hey, welcome back everyone. We are live here in London for HPE Discover. It's the Silicon Angles theCUBE, our flagship program. We go out to the events and extract the signal from the noise. I'm John Furrier, my name is Dave Vellante. Our next guest is Bill Mantell, VP and GM of HPE Enterprise, HPC. Welcome to theCUBE. And everyone, Bernard, President and CEO of Scality. Welcome back, good to see you again. Absolutely, good to see you. So the fruits of the labor of the relationship have been very positive. We've heard some great hallway buzz between the relationship and the HP and Scality. So congratulations everyone, we appreciate the time. So give us the update. What's going on? You guys are going to market together for a while. How long has it been since the launch? Any stats you can share? So the relationship started a few years ago, actually started in Europe and we now have more than 75 petabytes across several dozens of customers around the globe. But there was definitely an acceleration as we had a more formal partnership a year ago with HP servers in order to propose object storage in a software on industry standard servers fashion. And this week, HP Storage has also formally embraced the solution to complement the three-part portfolio which excels at performance-driven, latency-driven storage, kind of a tier one of storage, complementing it with a capacity-driven storage solution made of Scality software running on the Apollo servers of HP. Phil, talk about the growth opportunities because what we've been hearing is with the three-part and overall HP Storage line, store one all across the board, is that the seamless features that scale across a single architecture. Yes. And that's not just HP, that's also partners. This is one example. Give us some color into this opportunity from an innovation standpoint and the benefits of the customer. Sure. So what we're seeing is more and more that the amount of data coming down to this big data problem is challenging traditional storage architectures. And so the concept of Object Store which started in a lot of the big service providers is now starting to come down to a lower level. So it went from the cloud service providers into some of the telcos and now we're actually seeing enterprises that are needing a place to store all this data in a very fast but at the same time efficient way to get to it. I should say a very large and efficient way to get to it. And so that's what's driving this demand for object storage. Right. And so we've talked about object forever, right? And you see it in certain niches, photo sharing and things like that. Obviously see it in the public cloud. And now finally it's really starting to take hold. What's going on in the marketplace that people have reached this epiphany of, okay it's so much easier, so much simpler, it scales better, put get. This is an easier sort of mental model. But talk about sort of the maturity of Object Store. I think it's a couple of things. One object is a technology that's only what it is. And the adoption depends not only on the technology itself but what's around it. So it started with people who control the application stack 100%, could write the application natively in Object, big websites for example. And that's going to take time for every application in the enterprise to speak Object natively. So that's one aspect of the adoption curve. The way you work around that is to complement a native Object Store engine with a scale out file system which we did at Scality. So you don't need a gateway, you can present as NFS, as SIFS, and in the object world you can present as S3, as Swift, Native Rest, and so on and so forth. So that's point number one. Present the proper interfaces, take into account what the application really is, and give access to a legacy application, and give them access to the benefits of Object. Point number two, I think a key in adoption of Object is to apply the recipe of the web giants to the enterprise. Meaning software on industry standard hardware. And that's where the Apollo range of HP servers is magic because this 4U60 drive dense form factor is a perfect block for scale out architecture of that kind. Those servers by the way were not born for Object Store, they were born for big data, but they happen to do a magnificent job on a workload like object. And those servers are now available, and they're available not only to the big web scale guys, but to enterprise customers. So combine those two things, and that's where you get the adoption to start raising. And that was presumably the appeal bill of the relationship. What specific problem were you trying to solve where you sort of stumbled into scalability and said, hmm, this might fit? Well, it's again the demand for customers to be able to store all that big data that's coming down. And certainly as we've talked about, we've been selling servers into the service providers for a while in exactly this type of application. And so with the Apollo line, we actually brought what was largely a very custom engagement with service providers down into a portfolio, so that now more standard customers in the commercial enterprises and so forth can now enjoy the same levels of density, the same levels of cost value that they have in that. And so that's been a major change that we're seeing in the marketplace. And the big data context, so what's happening in the HPC world? Is it sort of the traditional HPC stuff that they're just now calling big data because it's always been big? People say big data, big deal in your world. Or is it sort of things like Hadoop and new frameworks coming into the HPC world? I wonder if we could talk about that a little bit. Let's try it. It's probably both from that standpoint. So there's always been a lot of big data in high performance computer workloads. Now what we're seeing is that that data becomes more valuable and more accessible than it used to be. And certainly with object storage over more traditional sort of tape siloed based infrastructures where it takes a fair amount of finite time to actually get to that data. Now, if you will, a lot of the data is at your fingertips for you to access. And what it does allow you to do is run more of what they call ensembles in the industry. In other words, you can take a large data sets over a period of time and run some of the typical type data analytics on top of them and get actually more value out of it. So a great example is we do have a customer that's done this in the, it's a Gen-Engine manufacturer. And they've actually looked at, they've run lots and lots of runs for some of their engineering analysis and then they use analytics to understand the difference between the runs. So you're actually starting to see that interesting blend of traditional modeling and simulation which is high-performance computing. And then on top of that, bringing in data analytics to sort of understand the data you have. And now because of the power of high-performance computing, you can afford to run hundreds to thousands of different runs on that. And then again, look at it from an analytics perspective. So this has really changed the science from that standpoint. So guys, tease the conversation and let's connect the dots to the big announcement around composable and synergy. I mean, everything's kind of fitting in like a glove's conversion infrastructure, hyper-converge, you got object, you got flash, all this stuff's kind of really kind of, there's a lot of stuff going on there for you guys to even keep up with. Never mind the customers, right? So they have to figure it out. So what's the, how do you guys tie into this composable? Cause I almost imagine with that kind of scale, this is a cloud-native kind of concept, scale up, scale out, I mean, on industry standard hardware is some benefits. The developers are going to take advantage of that. So we're seeing this composable message. Where does this fit in? What's your story with respect to that? Is there a fit? How do you guys talk about that? I think it's about bringing to the enterprise in shape of productized software and hardware, the same secret source as the web giants. And so from that standpoint, you're getting access to this large pool of storage where you can do HPC archive, where you can do big data analytics archive, where you can store any sort of large amounts of data you want to come back to afterwards. And you do it controlling that on premise, just like if you were yourself a Google and Amazon. That's kind of the contribution we're making to the whole story. So give them the data fast. And you, sorry John, but so you do that a little differently by bringing an appliance model to the table. Is that right? It's a software that runs on standard Linux, that runs on standard servers. We just happened the two of us to have a really good combination because the Apollo range of servers was there early on the market and kept its edge in terms of form factor, economics, performance. And so we happened to leverage the Apollo a lot for that leadership position reason. So technically it's a software running on Linux, running on hardware. It's really a software defined conversation. But from a customer standpoint, by joining forces we can deliver that as a very easy appliance. It closes that gap between, I want to be like that, but where do I put it? Exactly. Yep. And sorry John. And at the same time, all of the Hewlett Packard Enterprise portfolio is developed around the rest of the interfaces and these sorts of things. So they can all work together. You can manage them with a similar set of interfaces. So they'll all work together. So Apollo platforms that are working in the midst of our synergy platform. These all will work together with a similar, we all have ILO throughout, sort of allows us to manage it in a very similar way. So you can bring in, you can have a composable infrastructure and then have object storage as the object storage piece of that as well. And it all works together from that perspective. We're the big use cases from the object store you guys are seeing. I mean, we know the ones that are out there, not unstructured data is a good one. Was there other ones that you guys are seeing emerging? Any new information on how customers are using object? So the core use case remains active archive. So whatever organization who has a lot of research data or business data to store somewhere, behind big data analytics, behind an HPC cluster, or behind a regular, yeah. So that's the archival use case and that's probably still half of the market. What is emerging and very interesting as well is more active and performance driven services, video streaming. I mean, HP and Scality are helping RTL, a very big media group in Germany to do video on demand. And from that sole single storage pool, you can ingest and you can distribute your content which conventionally you were using several storage systems to do. So think of that as a really deep storage pool that can accommodate a mixed set of workloads. I think that's interesting. We saw each other the night at the storage party downtown London and you said, oh, we got SiliconANGLE theCUBE on Scality. You know, all kidding aside, this is a future move in collaboration. Video is the number one app. You look at consumer, Snapchat, Instagram, Facebook, and Facebook serving up billions competing now with YouTube. But that will roll right into the business video on demand, all aspects of, so interesting trend. Bill was mentioning that this started with service provider and goes to enterprise. Media are the first to pick it. We've seen that with a lot of joint success in the U.S., in Europe. And then everybody who needs a lot of video is facing the same challenge. We have a car manufacturer we're serving together, a European one who's storing seven petabytes of video crash test data on our kind of system instead of a legacy storage. Why? Because scale need for performance, need for future proof architecture. They are not video experts, but video is coming to their business a big way and pushing the handle on. They want simplification. They want scale. Absolutely. It's got to say it, it's less expensive too. I mean, it is. Sometimes you guys don't like to talk about that, but it's way less expensive. No, it's clouded coming for the enterprise, absolutely. All right guys, final question. What's your take on the show? You guys have been working together for a year tightly in a formal relationship. The split is over, congratulations. A lot of people excited. What is the thoughts? We'll start with scale, the new HP, your thoughts? Well, I think what I like in the HP enterprise for our own business is that we're coming from service provider going to enterprise and this week was full of enterprise customers. People who are in bank in insurance, in media, in auto car manufacturing, and are facing scale challenges with storage. And we're glad that the HP team are bringing these people to us and we're proposing the solution. And Tucker, who was just on the cube earlier there, having the same experience. HP's very partner friendly, they always have been. Absolutely. And this focus on enterprise is just really the right thing right now. Bill, your thoughts? Obviously people are pretty happy at HP feeling in the groove right now, I feel absolutely, yeah. So we've had a great show. It's our biggest discovery yet in terms of attendance. So it's a great coming out party for the new Hewlett Packard enterprise. And folks are excited in terms of our new direction and in the solution space and a lot of these other interesting markets, such as IOT, so there's a lot of buzz that we're moving forward. And so it's been all very positive. I got to say, I'm very impressed, fresh and relevant. Yep. And you got some cool factors going on. I got scale of the year and you got Docker. I mean, you got a lot of cool stuff going on. But the company too and more people show up. Got to love it. Guys, thanks for coming on theCUBE. Thanks for sharing your insight. We appreciate it. We're here live at HPE Discover HP Enterprise. This is theCUBE. I'm John Furrier with Dave Vellante. We'll be right back with more after this short break.