 Okay, we're back. This is Dave Vellante. We're live from IBM IOD at the Mandalay Bay in Las Vegas. This is IBM's big conference. We're talking about pure systems today, quite a bit actually. We're doing a number of drill downs on those and we're going to do some more here. But this is IBM's big software conference. Steve Mills gave a big keynote this morning. All the big customers are here. All the executives, the visions, the use cases, the partners, it really is quite an event for IBM. I'm here with my co-host. I'm Jeff Kelly with wikibon.org. So, we've got another great guest. We're going to talk about the pure systems family rather than kind of drilling down unnecessarily on any specific part of it. Jason Gartner, Vice President, Product Management, IBM Pure Systems. Welcome to theCUBE. Thank you. Yeah, so as I say, we've been talking a little bit about pure today. I've said, and I'm going to give you an opportunity to respond. IBM was a little late to the party with pure, but you came in with guns ablazing. Your colleague Nancy Person said, well, we had the mainframe. That's an expert integrated system. I said, okay, but there's a new type of expert integrated system. So, is it a fair comment that you guys actually came in a little late, but you did bring some additional value to the marketplace? I think it would depend on the workload. And I think that's really where we've been really focusing ourselves is really around the workloads that are associated to pure systems family. And I think if you look at it from a virtualization and an application space, absolutely not. I think we are absolutely unique in the marketplace. We have a value proposition that is very different. And frankly, I think it's unmatched. And I believe it will be unmatched for the next two to three years as we look out. From a data perspective, listen, part of pure data is Natesa. This is not late to the market. Natesa's been out there for years. I think that it's some of the others that are just starting to catch up because we're now taking and harnessing the expertise we got from the TISA and really expanding the workloads. And going after transactions as well as operational analytics with some of the new expertise. Yeah, I think a lot of this, the talk, it's kind of semantics first and market or not. But a lot of this is packaging, it's evolution of systems in general, the whole converged infrastructure play. Where's the innovation in this? Talk about that a little bit because you guys did bring some innovation, I think, particularly with the templates. Yeah, definitely. When I think of innovation around this, I see it in three major categories. One is just straight up and running. We are not a multi-day up and running system like every other system out there. You give us network, you give us power. We're up and running in four hours, whether it's application ready to deploy, fully virtualized environment, or it's data load ready. I think that is a tremendous amount of innovation is built directly with the manufacturing, built along the same lines and models of frankly our traditional AS 400. And doing everything from the storage is formatted, ready to go, direct out of the box. And that's one. Second is around development operations, deployments, the patterns, as we call them, and that's really around the application and the database patterns and how they can fit together and be able to be deployed. And then finally is around the thing that people really care about, clients really care about, and that's that ongoing maintenance and costs. And I think that's very unique to the marketplace is the savings that we can provide to our clients around that overall maintenance, tuning, optimization, and overall monitoring of system. Talk about why that's unique to IBM, because others will say that they're focused on that. Well I think in order to be able to do it right, we have probably over 60 to 70 years of data center experience. We have built a system that's not an island in the data center. It fits, connects, and interacts with the data center. And whether it interacts with our software or with our other hardware pieces, it's a seamless integration. And single fix for our maintenance, we can do it live on some systems like the pure application system. We can provide a single patch for the entire system. And whether it's single point of monitoring, single point of management, all from a very modern, new, unique UI that can manage the system from bottom to top. So what's, as you guys announced in April, what's the traction look like? I mean, essentially you're, this is a replacement for your blade server line, right? And you priced it very aggressively. I had lunch with Steve Mills at the announcement and I said, what's the premium that people have to pay for all this additional value? He said, there isn't one. I said, come on. He said, no, I review all the pricing. There is no, no premium. Yeah. Unfortunately, Steve does review all the pricing. But yeah, no, it's very aggressively priced. It's single package, single PID, single price. And, you know, around that, we don't piecemeal customers. We're not going after, you need this and this and this and this, right? When you go to buy a car, you're buying a car, not buying a whole bunch of parts. And so it was priced that way, it was packaged that way. And the momentum's been fantastic, right? Because we have that full spectrum and this conference has only really doubled my confidence in it because before we came out with infrastructure and application and application platform and really going after that cloud infrastructure and that application infrastructure. And now we're coming out with data to round out that family. And, you know, before we were very unique in that, that we had both aspects and now we've got the third to round it out. I see it as a three-legged stool and I think it's just, you know, it's starting to take off. Clients are very interested in it. We've got a lot of interesting responses and so we're pretty excited about it. So would you say interesting responses? I mean, so you're starting to have discussions about the fit, where it may or may not fit. Well, there's been some fit, but I think the interesting response what I was referring to was new clients' IBM. This is not, this is not- New IBM footprints. New IBM footprints. This has not been going after the traditional IBM footprint. Now we've been talking to obviously the traditional IBM footprint, but we've sold the system into Inner Mongolia as a cloud deployment platform, not a whole lot of existing IBM customers there. And a lot of the interest is really around some of these new ones because they are frankly getting it fed up on either, you know, the pricing. They're getting fed up on some of the quality that they've seen and they're looking for someone who's a real partner and who has the history and the proven technologies to be a partner in what they want to do. We're undergoing a transformation in the IT. That's what is happening. So how does peer systems impact the data center professional who's, you know, you've got all this experience at IBM. You've seen the ebbs and flows over time. What does this mean for the data center pro that the admins that are working in the data center every day? Yeah, so to me it's about getting smart people to work on hard problems. The CIOs, they want to drive more innovation. They want to drive new applications, new businesses, new markets. They want to change the IT economics. But they can do that when their smartest people are constantly doing a lot of the mundane tasks that are happening out there, whether it's around maintenance or management. For example, a very easy example is on a small system of a peer application, just 239 endpoints of maintenance. That would, what would you would have to do if that was your own system? And that would take you six months because there'd be a whole bunch of dependency management. You'd have a massive spreadsheet. Now today, one patch and it can be done live. You know, and it can be done in hours, not months. And so you're really given that time back so that they can drive new applications and be able to drive new, go to market routes for their CIOs. So how is that dynamic changing the operational structure in your client organizations? Yeah, I think there's a little bit of, you know, redux happening here as we've, you know, gone from, you know, these integrated systems of the mainframes and as the client server expanded, you know, the windows explosion, which really, you know, created this, you know, mass market of systems. And then, you know, really it's all been coming together with these integrated systems. And it's around who can provide a system that can be administered by, you know, fewer people and with less time. Because really what they want to be able to do is be able to take that savings that they have and go after new markets. And yeah, there is definitely a transformation that's happening in the IT culture. You know, especially as you start, you know, bridging across network and storage and compute and middleware. But frankly, they've been asking for us. You know, the CIOs have been saying, I need this. And so there is a transformation that's happening. And there is a little bit of growing pains, I guess, with some clients as they go through this, which is maybe why it's interesting to see some of the newer clients, especially in some of the growth markets, are embracing some of this much quicker. They may be some of the more traditional, huge, you know, pillared IT shops. Right, they don't have that legacy to overcome. They don't have that legacy to overcome. But organizational change is fundamental to cloud. It's fundamental to really the cost savings and we need to be able to drive in a lot of these organizations. So is there a proportional relationship between the guys who are adopting cloud and the guys who are putting in this expert integrated systems approach? They are tightly linked. They are absolutely tightly linked. You know, whether it's from cloud, because cloud is really around the cost savings, the standardization, it's around automation and consolidation. Well, that's really what Pure Systems is about, right? It's around standardizing your application and deployments through patterns, standardizing your infrastructure through infrastructure patterns, or standardizing your databases. It's around consolidating them in single systems, driving higher utilization rates. And, but all of these have the same challenge. It's around organization. Have you, I mean it's been a while now, so what's it, six months? Yep. As you launched, do you have enough data to talk about how Pure has changed the expense structure within your organizations? So we have, so we've done, you know, with a lot of these clients that have been looking at the system as well as purchasing. Obviously that the modern IT shop today is very specific on what savings it expects and it holds us accountable to that. And I think when we first launched, we really came out thinking up and running. We were thinking being able to roll out new applications much more quickly. And that has definitely transformed to be true. But the CIOs, when they look at some of the savings that they have and they look at that, they say, great, okay, it's improving my dev ops. I can now spin my applications much more quickly through dev test production. But then they look at the numbers that they feel they can save on an ongoing maintenance basis and that gets them excited. I think that's that core nugget that they're looking at and going, wow. And these are from existing application environments to something which is managed here. We're in the neighborhood of 70 to 80% savings in that overall management maintenance of it through the pattern-based approach and through that single point of maintenance for the system. I mentioned dev ops. I'm interested as to my same question that I asked on cloud. Is there a proportion of relationship between people that are now doing dev ops? Because you've so simplified the infrastructure, are you seeing that trend toward dev ops, the intersection of operations and application development? Yeah, absolutely. I think when you look at, so we said we simplified the infrastructure, we also simplified the application platform. So we provide a set of shared services for the developers. We provide the ability to self-provision everything from a three-tier web application to a Cognos deployment, a business process management, a portal. We've got over 190 ISVs that have built applications and application patterns. So now you can download them off the catalog, deploy, test, take that exact same pattern and go test dev test production or dev test staging production with the exact same configurations. How many times you see a client saying, wow, my test worked fine but my production is sick. What's going on? What's the difference? What's changed? But now you create that single point of truth of that deployment in there. It's huge on dev ops. They take them from three months of provision of three-tier environment to now they can do it in less than 20 minutes. I think it's huge. Yeah, the problems are cute. When you throw code over the wall, hey, it was working fine when I said, wow, you were running that code in your laptop and now I got to put it in production and then they start hacking away at the code. Well, you broke my code a minute. It's just the nightmare and then you get down to the last minute and it's everybody spending. That's right and it's all hands on deck and the studies we've done is just looking at how many projects are late, configuration, management, tuning, and connectivity around to your data center because you might work on an application but then when you want to put it in production it's got to connect to everything. You got to connect to your security. You need to build an elasticity on it. You need to build in the ability to be able to grow the system very easily and quickly and that's really where we've seen a lot of advantage. Jason, does expert knowledge eliminate the need for certain professional services? No, I don't think so at all. I think what it does is it allows you to be able to focus that professional services on really the value for the client, right? A lot of banks don't want to be an IT shop. They want to be a bank, right? Retail doesn't want to be an IT shop. They want to do retail and really what they're looking for professional services is for that industry based knowledge to be able to really focus what they want to be able to do around their industry and create value for that industry and create that competitive advantage. They don't want them spending time integrating systems and setting up and racking and stacking and cabling. That's just not their core competence. So your premise is that they're shifting spend from that racking, stacking and cabling to higher value functions? Yeah, and I think from a professional services perspective or a partner perspective is now they can provide higher value services with greater margin. Are, yes, absolutely. Partners, I would think, love that. Certain partners anyway, the ones that get that model. But I want to push at that premise a little bit. Are you seeing your clients gain share with the CIO or are you seeing the CFO drop it to his or her bottom line? So it is going both ways. That's why we tend to talk about hard savings and soft savings. Because even if you save people, what are you doing? Maybe they don't have to work as often in the weekend and we actually hear that a lot less. It's that I'm actually able to focus on providing action, doing things right as opposed to spending my weekends, I get them back. So there has been a little bit of both ways. I get to spend time with my family in the weekends. I imagine how innovative. Yeah. It sounds like it's certainly the pure systems approach by saving time and money that feeds the ability to do some of the big data things that we've been talking about at this show as well. They're kind of self-reinforcing approaches. When you look at most data centers today, between 16, 70% of their budget is spent keeping the lights on. Very few, only one in five can actually spend more than 50% on innovation and new projects. Well, that's the competitive advantage is new projects. So if you can shift that money to new projects, to new markets, the CFO would love to be able to grow that top line. And that's really where we're seeing a lot of that shift happen is into some of these new applications, new projects, new insights in their business. How are people migrating from their legacy environments to pure? So the system is 100% open, right? So whether it's from the database from pure data or whether it's from the application space or from Flex, we provide that complete openness, right? We provide Linux based solution, we provide power-based solution. You have an existing application that can run on that. If it can run on an existing legacy infrastructure, it can run on any one of our systems. And so I think there really isn't, that's why I do like the word evolution as opposed to revolution because you're taking an existing application or existing way of doing it and evolving it to a higher form. So most of these integrated stacks though are highly homogeneous. You got x86, you got power, you got FC, you got Ethernet, you got Infiniband, you got different virtualization, you got virtualized, you got physical. You got sort of all of the above. Can you attack that labor problem with such homogeneity? Well, I think that's when it comes down to that single point of management, maintenance, monitoring, because I don't want you to know what's inside the box. I don't want you to know what's inside, you know, it's like, I don't want to know what's inside sausage. I want to know that I'm eating sausage. And that's really where, you know, I think that true value differentiator is because we don't need you to understand that because we deal with the management, the monitoring and everything for you. And we've been able to prove that out with a number of clients that have been running other operating systems, that have actually moved operating systems, they weren't part of their standard. And if they've been able to do it, I know that a lot of others can do it. So you threw a curve in my analysis because I drew a spectrum and I said, okay, at this end, you got the reference architecture, you know, a lot of choice, at this end, at this end, you got single block of infrastructure, any color you want, as long as it's black, and that's the two ends of the spectrum. And essentially what Pure did is came and said, yep, we got all that. We got single skew, right, if you want that, and we got, and we have choice. Yes. How are you able to do that and everybody else has to play in either end of that spectrum? Well, I think where it comes down to is frankly leveraging that build your own infrastructure and the management associated to that. Not just using straight on commodity hardware, but really going after and taking and building on that management platform, and then using it as a common management platform. We've spent, you know, frankly, over a million hours building a common management platform around this. And I think that's where the uniqueness comes into play because now we can have that system make some choices, be able to do virtualize, be able to do application and be able to do data. And I think it's really around that management platform which is unique. And as that family begins to really solidify around it, it's going to be that management platform which is where our administrators are going to spend time, our developers are going to spend time, and frankly, where we're going to see the savings. So we're out of time, but I have to ask you the storage question earlier, I guess, that nobody wants to talk about storage, but storage can be up to half the budget. So what's the storage angle here? So we do have integrated storage. We've got both spinning disk and solid state. And, but really what we've done with around our storage is really optimized how much, what kind, and how it's connected. Whether it's application, warehousing, database transactions, it's slightly different based upon that workload. We've integrated, you know, fairly innovative technologies. In that, we've integrated an algorithm into the system so that they can monitor the spinning disk, be able to recognize hot spots within the spinning disk, migrate it to solid state automatically. And what that does is it increases the performance by up to 2.3 acts on an existing application that's just running. And I think clients are seeing that and they're just going, wow, and the amount of administration around that is zero. It's automatic. Jason Gartner, thanks very much for coming on theCUBE and sharing your perspectives, taking us through the lineup of Pure Systems. Congratulations on the launch. Good luck and appreciate your time. Thank you very much. Appreciate it. We'll be right back with our next guest. This is theCUBE. We're live from IBM IOD, right back.