 Live from Las Vegas, Nevada, it's the Cube at IBM Edge 2014. Brought to you by IBM. Now here are your hosts, John Furrier and Dave Vellante. Hey, welcome back everyone. We're live in Las Vegas for IBM Edge. This is the Cube, our flagship program. We go out to the events and extract the signal from the noise. I'm John Furrier, the founder of SiliconANGLE. Join my co-host, Dave Vellante, co-founder of Wikibon.org. Our next guest is Chris Saul, marketing manager for StoreWise at IBM. Welcome to the Cube. Thank you, John. So IBM Edge is rocking. StoreWise brand has been there in the middle of all the action quite some time during this transformation. Yes, we have. So talk about how that relates to you guys and what's it like here at IBM Edge for you guys. Well, you know John, it's been fantastic so far this week. We're using Edge to announce big improvements to our StoreWise family and to our sand volume controller system. And, you know, we're announcing new hardware that uses new technology from Intel. Intel's Quick Assist technology. IBM is actually the first vendor to be delivering this new Intel technology. What it enables us to do is to improve the performance of our real-time compression. And when we do that, we can deliver as much as five times as much data on the same storage systems with the same or better performance than people were seeing on traditional storage systems. One of the themes that's clear here at IBM Edge and certainly not so much in the other shows at IBM, but here at Edge, because you're in the infrastructure side of it, is the integration of the technologies at the chip level. You're seeing, you mentioned Intel, right? So, I mean, I know Intel is working on a lot of a bevy of products where, you know, security, stop getting embedded into the chip level. We heard from Mike earlier, Mike Coon, about talking about some of the differentiators around the storage side, right? So, what does that mean to you guys? Because you guys are doing compressions, heavy-duty stuff. What innovation is going on inside the hardware that allows the software developers to write better code? Well, what's happening is Intel has produced a new set of chips, which they codename Colleto Creek. And what Colleto Creek does, it's a specialized chip that's dedicated to compressing data. And so, with this system, with the new hardware that we've integrated, we can deliver as much as ten times the performance for compressed workloads compared with what we were doing before. And when you... Ten X. Ten X. That fits the Larry Page model. Ten X. I saw a book in the airport. Ten X performance. That seems to be the magic number. Yeah. I think, you know, a number like that really gets people's attention. And when you combine the Intel hardware with IBM's real-time compression software, which uses that hardware to deliver customer value, and that's really what we're all about. And that's why we're saying, with the new StoreWise Family hardware, you can be thinking about storage for half the cost of what you might have been looking at traditionally. I've always loved the StoreWise brand. Dave and I, we've had been talking about, since the acquisition, that little tuck under deal. Love the name. But give me an example of that Ten X, because that does get your attention. That is a moonshot level kind of benchmark. What use cases are you guys tackling with that kind of power? Well, really, you know, what it's all about is pretty much any workload. We have a customer who is speaking with us here at Edge, Palafone, who is one of the largest mobile providers in Israel. They were talking about their experience using real-time compression, using our sand volume controller technology, using our XIV storage. And when they turned on compression, they were able to get 10 times as much data in the same storage systems that they were using previously. And they got better performance out of the system as well. And funnily enough, the technicians were talking on the video about how, you know, tricky it was to explain that to their management. Who, you know, were like, how does this magic happen? How can you get, you know, that big of an improvement and you get better performance as well. So, you know, that's what's happening in the real world. So Chris, I got a question for you. Big companies, large storage companies. Why can't they build from the ground up organically? Why can't they innovate, build products that can ship tens and thousands of thousands of systems? Why can't they do that? They can, actually. So there's a meme out there in the business that large companies can't innovate. And so, I mean, look at it. EMC's criticized by that, about that. Certainly HP had to go out and buy 3PAR. Oracle is never really in the hardware business. But, you know, it bought sun. So, SVC is an organic innovation. Why is it, of course, I was joking before. Why is it that IBM was able to do that? How did that come about? Was it a skunkworks in England? Was it a gleam in somebody's eye? Was it an R&D project that somebody tried to kill 10 times but finally hit the light of day? You were there in the early days. I was. And actually, you put your finger on it right there at the end that the way Sandvolume Controller and then more recently, store-wise, really happened. It started as a research project. And so, a big part of the reason why IBM can be successful is the enormous investment that we make in research. And so, back in 1999, we had a team of people in Almedin who developed the idea for what became, in 2003, Sandvolume Controller. And then we have been delivering improvements to Sandvolume Controller, new hardware platforms, and then more recently, the store-wise family, built on the same technology since then. And I think, you know, you were talking about innovation in large companies. And one of the things that we've aimed to do, and we've consistently done, is to look around IBM and look for technologies that other teams might have been developing and see how we can include them into the system. So, for example, we stole the user interface from XIV, which most people acknowledge, you know, is the best interface in the industry, and that's now part of the interface for Sandvolume Controller and for store-wise family. We took the easy-to-hear technology that started in DS8000. We built that into the system as well. And then more recently, we've actually started looking outside IBM. And so, we integrated technology from Bridgeworks called Sandslide that optimizes the use of network technology for remote replication to reduce cost. So, it's this attitude of saying, we'll look for the best technology we can, you know, whether it's within IBM or whether it's coming from outside IBM, and integrate that into the system. I mean, IBM's had a sort of a very interesting history in storage, obviously. Sort of ebb and flow, and sort of big investments and pull back on those investments and then started to make those investments again. And you're now starting to, I think, see the results. You can't just turn on and turn off and then turn back on. IBM did in storage, but the sustained investment is now that you're seeing it. Yet, SVC was born in that climate of uncertainty and it succeeded wildly. What are the numbers now? I mean, you guys always talk about more than 10,000, and I think that's probably still the number, but what about the V7000 or the store-wise family? So, as you say, we've sold more than 10,000 sand volume controller systems. That must be way more. It's one of those things that's kind of tricky to count, so we don't count it too often. With our store-wise family, we've sold more than 55,000 systems. We've sold more than 130,000 enclosures of the drawers that the system's made up of and we have more than 1.4 exabytes of capacity under management with that system. And, you know, if you look across our entire worldwide installed base, so all of those systems across the world, we're delivering better than 5.9s availability with those systems. So, we're delivering a lot of capacity, a lot of performance, but doing it in a very reliable way as well. Yeah, now some people might say, well, you know, you went about store-wise and you brought in real-time compression. That's real kind of compression. That's all true, but that is an organic invented in IBM system. Period, correct? The system is invented. And, you know, you said earlier about, you know, was it a SkunkWorks type technology? Yes, it came from research, but for a number of years it was bordering on SkunkWorks type things, but what I think one of our great strengths is that the key architects of the technology, the people who developed the technology back in 1999 are still leading the technology development today. So, we have these, if you like, guardians of the architecture of the system who know, you know, why we designed it the way we did and who enable us to continue to deliver it in a very high quality manner and have really shepherded it through its life. Now, it's interesting because the ITC technology has become an underpinning of IBM's software-defined strategy, as well as its flash strategy, which is quite interesting because many flash systems, including, you know, TMS when IBM bought it, lack a robust stack. So, you said, okay, just put it behind an SVC and voila, instant stack. That's right. So, talk about the importance of stack, why it's so hard to build a hardened stack and why it takes so long. It's tricky to build a hardened stack for a storage system because you have a number of different things that the storage system is trying to do. So, we're trying to deliver performance with caching, for example. We're trying to deliver different types of functions. So, perhaps local replication and remote replication. We're trying to deliver functionality like thin provisioning. We're trying to deliver automated tiering like our easy tier technology. We're trying to deliver an influence on the others. And so, it's very, very important, we believe, to have a strongly layered architecture so that you can keep the different functions separate and so they don't interact with each other. The result of this is that with Sandvolume Controller and with StoreWise Family, you can use any of the functions in any combination because our architecture keeps each of the functions separate. So, if you want to do replication between a thin provision volume and a non-thin provision volume, you can. If you want to make a copy from a compressed volume to an uncompressed volume for some reason or the other way around, you can do that. And if you look at a lot of our competitors, you find that because they're less careful about protecting the architecture, because they're less careful about separating the functions, there are restrictions that we don't have with our system. So, you can't use a thin provision volume if you also want to do snapshots and those are the sorts of things that make a system harder to use and we believe that we're serving our customers better when we make a system that's easy to use, when we make it flexible and so they really don't need to spend a lot of time worrying about how the different bits of technology work together. You're essentially saying your code base is not a hairball and that some others are. That's essentially what I'm saying. Because I was going to ask you, because people talk about it's old, it's an old system, it's going to bolt on, you're describing a situation where this is advantageous because it's mature and hardened, but it's designed in a way that is not all clouded and intertwined and constrained. That's exactly right. Let me give you an example of that, that when IBM acquired the store-wise company, one of the reasons for bringing store-wise the company into IBM was to integrate the real-time compression function. We were able to do that very quickly with sand volume controller because of the architecture that we had, we could just plug the store-wise technology in and we got real-time compression. Similarly, I was mentioning the Bridgeworks technology that improves performance for replication. In the same way, because of our layered architecture, we could just plug the technology in very easily and add to the system and give our customers value much more quickly than we could have done otherwise. If I interpret that right, you're saying once you started the project, you were able to integrate it very quickly. But it took some time to get it into the queue. Was that just because you have so much stuff in the queue? Was it because the sales guys were maybe afraid? We don't want to cut the price. We don't want to sell less storage. Was there some of that going on? That's a very interesting point because when you talk about storage efficiency, whether it's compression, whether it's thin provisioning, whatever it might be, there's always a little bit of attention with salespeople because the immediate reaction is, it's not storage so that seems like a really bad idea. But the flip side of that is that whatever it is you are trying to sell becomes easier to sell because it's now much more appealing to your customers. It's much more price competitive and so people are going to want to talk to you about that. I mean, everybody wants to find out how you could save cost in IT. Plus the market is, we've seen time and time again, the market's elastic. Cut the price. People almost invariably always buy more. That's certainly true. And now that there's more dates available that people want to store. All right, Chris. Thanks very much. Chris, great to have you on theCUBE. StoreWise, this is theCUBE, live in Las Vegas for IBM Edge. We'll be right back after this short break.