 Live from Las Vegas, Nevada, extracting the signal from the noise. It's theCUBE, covering IBM Edge 2015, brought to you by IBM. Hi, buddy, we're back. This is Dave Vellante with John Furrier. This is Edge 2015, and we're live here. This is theCUBE. Steve Sibley is here. He's the vice president of Power Enterprise System. Steve, great to see you. Welcome to theCUBE. Thanks, Dave. Nice to be here. So, John, you remember Edge, the first Edge in Orlando three years ago was really a storage show. So, they really expanded that, Steve, and to including Power. You got to be pleased about that, Zee, middleware. So, what's that mean? Yeah, it's great to be a part, right? And as we look at putting more and more of our systems together, driving more synergies across our entire portfolio, storage systems, even our middleware, right? Being able to drive additional value, it's great to be part of the same conference. So, we just heard from Ken. He was talking about the Open Power Initiative. Your role is to focus on bringing power into the enterprise. What are enterprise customers doing with power and sort of what's new in the product line? Yeah, so a tremendous amount, right? Obviously, our enterprise-class systems have been kind of the core of our power systems business for many years. We have clients that depend on power as their system of record, right? The system that runs their business that they count on every day, 24 by seven, 365 days, to be up running, managing their data and executing many of their transactions. That's what our enterprise systems do. So, I had to ask you about some of the characteristics of the enterprise, and Dave and I have been covering them. It's our sixth year doing theCUBE, and it's always been scaled up in the enterprise. A lot of general-purpose computing, you guys have mainframes on the high end, and then you have software and systems and the other, but the trend of scale-out has been a very interesting dynamic. So, what's the blend between scale-out and enterprise-grade? The scale-out kind of feels like open source, commodity hardware. You guys aren't in that business anymore. Power has an industry backing behind it. What is this new dynamic, and give some examples of where it's working and what are the challenges are? Yeah, I think one of the key things to recognize is much of the scale-out growth has been driven not only from the commodity hardware, which certainly has been a contributor, but also the application development from the open source community, right? And so, our focus and drive into scale-out has been around openness, new applications, expanding the ecosystem for power, right? We've put some great new products in place that have done extremely well over the last year. Scale-up provides additional value to clients that are primarily running, again, their systems of record applications, right? And now we're actually bringing some characteristics of both of those together. We recently just announced the PowerEat 850, right? This week, right? Where we are bringing many of our enterprise capabilities into that four socket space, which has some scale-out characteristics as well. You guys, it used to be the old speeds and feeds. Hey, I want this box, it's got to have all the ports and speeds and feeds, and that's kind of a scale-up model, you know, hey, and buying gear. But now you mentioned applications. Everyone's talking about workloads. Talk about how that's changed, the customer's purchasing power, and developers, et cetera, because now workloads, running workloads on, whether it's cloud or systems, is like the top priority. So how do you guys talk to customers with that context? Yeah, it's actually quite interesting, right? Because the systems themselves, people still need the performance, right? We don't talk about the speeds and feeds so much anymore because they are focused on the workloads, but they want those workloads to be performing outstandingly well. In fact, in today's weightless world, right, we are delivering better capabilities so for them to get insights faster. So the performance is still important, but it's really interesting what they really are looking for is the resiliency in their infrastructure, right? The ability to run their applications that are really important, right? But also flexibility, things like our capacity on demand that we have on our scale-up systems, our enterprise-class systems, give them the ability to respond instantaneously to changes in their business. They can add capacity without having to re-architect, without having to bring in new systems. They just turn on more processors, more memory. They instantly have more capability to meet those new demands from their clients. And do you see that requirement in something like a four-socket system, a capacity on demand? Yeah, so it's really interesting. When you talk about today's E850 as a four-socket system, there is a tremendous amount of capacity and performance. This has been a sweet spot for power for many years of mid-sized companies that are running their business on this kind of capacity system, and they want that same flexibility, that same kind of efficiency as our larger customers as well. I just got a note, John, I've been saying that this is our third edge. This is our fourth edge. We've done every edge, right? The first one was four years ago from that. Anyway, sorry about that, everybody. So, and then, where's the E880 fit? It's a much larger machine. What are the workloads for that system? So, when we talk about the growth in system of engagement kind of solutions and applications from open source, that doesn't mean that our clients aren't growing their systems of record as well. In fact, we have many clients who have been asking for and waiting for the E850 or the E880 with its scalability to 192 cores for at least since we started rolling out PowerA. So, we have a lot of demand for people who need more and more data and capacity on these systems to execute more transit actions. And the 16 terabytes of memory that we are putting on this system, the amount of processing capability is absolutely necessary for some of our larger clients to be able to run those workloads. And you've got a lot of roots in HPC. You've got a stronghold there. Drive and John and I talk about all the time how the sort of characteristics of big data, very similar to what's going on in HPC. Talk about that a little bit in terms of a tailwind for you guys. Yeah, so analytics is taking a lot of different areas, there's certainly in the high performance analytics or high performance computing area, there's both kind of the clustered scale out kind of workloads that can be optimized. And certainly with what we're doing around open power, bringing NVIDIA as a key member there to deliver new capabilities is important. But there's also some analytic applications that are emerging that need the SMP capability that a scale up system has. Some unique areas that we're starting to see in financial services and other places where you need not only that ability to scale out workloads and send queries over multiple systems, but actually a high performance, huge memory capacity system can deliver new levels of analytics in those areas. Was fraud detection an example of that? Yeah, fraud detection is certainly wanting to be able to understand immediately whether or not this is really you in a fraud, a credit card kind of situation so that you know without a doubt that you should execute that transaction. We talk about real time all the time. You use the term weightless insights. That's kind of, is that your new term? Yeah, so a new thought process, right? And a new perspective as we go through life, right? As much as we expect things on demand, right? And now with our mobile devices and ability access so much, we expect things to be there at our fingertips. We still actually spend a lot of time waiting. We have many clients who want their analytic reports instantaneously, immediately, not two or three hours from now, right? And so we're going into a world where you shouldn't have to wait, right? You can't wait to do analytics across fraud. You can't wait to do analytics as you're looking at things like security exposures or even potentially terrorism events that may occur. You need to know now immediately that you have an issue and that how to address that issue. And it's consistent with what we've been covering. We look at Amazon, for instance, in the cloud, commodity, compute, compute, it's going to zero, but yet the application focus you guys focus on really is highlighting the shift, that the table stakes is power. We need more high performance compute. That's like not going away. At the same time, you got to give agility to the applications. That's the shift that you're seeing. That's kind of what you're going after. Yeah, it really is a mix of this idea of needing the applications, needing the immediate response to those applications, but being able to go and address those new markets immediately from an agility standpoint. People use the cloud, yes for the commodity, economic efficiency, but really it's more than that. It's the ability to create new applications without having to spend months setting up your infrastructure. With the agility that we provide on our enterprise classes with capacity on demand, you can get that same kind of capability where you turn on new capability immediately without having to wait because you've already established the infrastructure. I've never interviewed a customer. We do a lot of customer interviews here in theCUBE that said, I want less compute. It's always, it's going more and more because quite a lot of the breakthroughs out there right now are new things people are discovering, whether it's security threats or something else. So this seems to be the applications that are dictating to the infrastructure. Yes, absolutely correctly, right? And the fact that these applications not only need to respond quicker, but need to be able to access more and more data is part of what, when we talk about, when we talk about our systems being designed for data, whether it's our scale out systems that scale to two terabytes or these scale up systems at 16 terabytes, being able to feed the processor so that they can respond to the application requirements is absolutely critical. Okay, so I've got to ask you the question for the customers out there that might not be buying IBM in this capacity. What's some of the use cases, the pattern that you see out there where you guys would be a great fit? Is it certain verticals? Is it use cases, certain workloads? Could you share some examples? Yes, well, we're certainly in most of the major verticals, but fairly dominant in things like finance, retail, telco kind of environments where they have many users, a lot of data that they're trying to pull in, right? Those are where our scale up systems that have these tremendous capabilities fit best. But really, as you look at some of these new applications and new workloads from a use case standpoint, some of the new unstructured data environments, some of the unique things we're able to bring together with open power, right? With our Redis Labs application and solution for in-memory, no-SQL database, right? Something that no other platform can do with the efficiency and effectiveness that we do. Yeah, we've seen some companies had great specs on paper, but missed the market because they're not aligned with some of the trends. So I want to talk about the developer angle because you mentioned Redis, Linux. The developer environment in the enterprise is also transforming as well. So how do you guys fit with that emerging pressure that the customers have to hire more developers? Certainly, it's legacy programmers. I mean, we were interviewing IBMs. There's still COBOL out there in the mainframe, but Linux, as it becomes more and more prevalent, Redis, these are cloud-like technologies, hybrid cloud. There's kind of a convergence there. What is your fit with the developers and how do you guys work with the customers there? Yeah, no, that's a really good observation. And as we've driven the shift, right, and the focus on Linux and the open source applications, teaming with Red Hat, Canonical with Ubuntu, Suce, right, to bring their environments fully with little Indian Linux onto our platforms, it really is a focus on new applications and developers. The open source tools like Python, the open source management environments like Chef and Heat, right, to orchestrate and deploy those applications are absolutely critical. And so we are working to not only make it easier for them to access capability, putting power in software as an example, but also how do you attach and drive your environment differently? How do you connect those systems of engagement, new applications that can be developed on PowerNow, run across any of our portfolio now that we support little Indian Linux on PowerVM, but also connect it to your system of record applications, either co-locating it for higher performance or connected in a hybrid cloud with things like our Bluemix security connect solution that we team with our middleware team on. You feel good about the developer positioning with? Yes, absolutely. In fact, we have brought a lot more Linux applications to the environment, and we're getting the attention of the corporate developers as well to develop applications on Power, on Linux. Well, it's interesting what you guys have done. I mean, you're kind of the underdog now of the systems business, and you've rewritten the playbook, talking about opening it up, building out the ecosystem, little Indian, just sort of a very smart move, right, because you're opening up applications. So that had to be an interesting planning exercise internally at IBM, and it's starting to bear fruit. Maybe talk about that a little bit. Yeah, absolutely true, right? When we took a look at where the market was going, we knew open had to be a core part of our business. Open now infuses everything we focus on across our systems. Not that we are not continuing to invest in AIX and IBMI and our core platforms in business. In fact, those are still critical for our customers, and our focus is on extending those with open technologies and open capabilities, right? So whether it's open power foundation and the over 120 members we have there that are infusing new technologies and innovation in power, or our open stack initiative in the fact that all of our virtualization management and cloud management is moving to this new model of open management and innovation that's occurring, or the open source applications and tooling and solutions that come out of Linux and other scripting languages, our focus as being the most open systems provider in the industry, and we think we're accomplishing that with power. So we have to go, last question, goals of the business, what are the goals of the business? What milestone should we be watching as independent observers, as progress and indicators of success? Yeah, so clearly our primary focus is on helping our clients with their innovation, helping them grow, right? And we'll see that in our business through their investment. And so seeing us return to growth as we did in first quarter, but continuing to see the growth in our ecosystem around these new open technologies and these new open capabilities, and our ability to bring those technologies to our clients in new ways. We're able to do that now with what we're delivering with the newest NVIDIA GPU processor, with these solutions as we build on the ecosystem stack around data and analytics with NoSQL or Hadoop type of solutions and those, and so I think really the breadth of number of members in open power, the breadth of Linux solutions, but also how we package and deliver those clients so that's reflected in their investment in power and obviously the growth in our power business. Steve Sibley, thanks very much for coming on theCUBE. IBM power, power play, making some new moves, opening up the platform, building out the ecosystem. Thanks very much, good to see you. All right, keep it right there, buddy. We'll be back with our next guest. This is Dave Vellante and John Furrier. We're live from IBM Edge 2015. This is theCUBE, we'll be right back.