 Hi everybody, this is Dave Vellante of Wikibon.org and this is theCUBE. We're here at 590 Madison Avenue at IBM's Flash ahead announcement, Steve Mills and his crew just announced that IBM's going to invest a billion dollars in Flash. They're going to set up competency centers. It's clear IBM after its TMS acquisition is all in. We're here with Karim Abdullah of Sprint. Karim is a TMS customer, IBM customer and a practitioner of IT and of course of Flash. Karim, thanks very much for spending some time with us. You're welcome. We've been talking, we had dinner last night and we just came off a panel here at IBM. Tell us a little bit about, again, how you're using Flash inside of Sprint and what it means. We're using Flash in the IT data center space to address many performance issues and storage, most notably in the call center area where we have to satisfy call center customers. We have 121 call centers across the globe and we have files where they need ready access to respond to the customer's queries and issues. We were having a significant performance issue there and by adding Flash memory into the environment, we have boosted that response time significantly. In fact, it's led to one of our recent award where we are actually ranked number two or three in best customer care now in our industry and part of it has to do with the way our customer care agents are able to respond to the end customer and provide a resolution to the ratios. Can you tie that, can you actually tie that benefit to Flash? Yes, yes we can. Really? The fact that they can readily access their data to provide a response to the customer and then we measure the customer care agents' ability to solve an issue with a force or second call and the response time they take to do that. So if I had asked you two, three, maybe four years ago, if you're going to use Flash, you would have told me it's too expensive. Too expensive. What's changed? The fact that today, you saw the announcement it's around 11 or $12 per gig. That's probably where tier one storage was two years ago. So it's the transition from that at a pricing point does not make a big challenge anymore. The problem is you're able to show the customer a cost benefit by able to respond to a customer, retaining your customer and at the same time giving them the data they need. So we talked about this in the panel a little bit. I wonder if you could just add a little bit more color. You know, there's this scenario where you're popping into tier zero and that's what you have done. That's the Flash tier. But there's also a discourse in the industry where people say, all right, we're going to have all active data on Flash and then the rest of it is going to go to a bit bucket. You know, slow, saturized. You don't see it that way, do you? No, we don't because we've part of our industry requires us to keep data around for too long. And most of it has to be accessible. So you need to have tiered storage, meaning high end and low end. So we don't see low end storage going away too fast or any time soon. In fact, we still have tapes in our environment. So yes, there's a role for TMS type storage or Flash memory or Flash storage. There's also a role for tiered storage, tier one, two, three, and so forth. And we see them all playing together nicely. You put a lot of storage behind the SVC, including the Flash, correct? Yes, we have. That's IBM Sand Volume Controller, okay, which is essentially a storage hypervisor. Do you put your tier one behind an SVC? Yes, we do. We have tier one, two, three, and now zero behind SVC. Okay, so, and you see that as evolving just to incorporate that tier zero. And then how do you get data off of where it doesn't belong? Let's say the data becomes stale and needs to get off a tier zero and go to the bit block. Well, that's the beauty of having it behind a virtualization management structure. This way, you reduce the number of hand motions, if you will, or operations and intervention by just having SVC manage that motion for you. It's really automated. Okay, and so you're a multi-vendor shop, like most customers. You chose the IBM Texas Memory Systems system. Why and how they doing for you? Number one is interoperability. It was easy to deploy and it just looked like another storage node in our infrastructure. They were out of energy. We looked at, they were a level of difficulty into integrating them in the environment. The other part was getting favorable cost points and with TMS, we were able to do that. So that's a block-based device that you just drop in. It looks to the applications like any block-based device. So that really simplifies your life. Yeah, okay, and long-term, where do you see this whole flash game going? How is it going to change your infrastructure and your applications and ultimately your business? In a number of ways. Reduce footprint, reduce overall operations costs and not just performance to the end customer but the ability to troubleshoot issues faster. Because we're approaching real time for most things in the IT environment. Not just solving but providing troubleshooting. And we talked in the panel last night, you actually see the energy bill. You're one of the IT guys that actually pays the bill so flash obviously makes a difference there. Absolutely. My commitment is to reduce power consumption in a data center one to over a month. And that reporting goes up to the Department of Energy actually because we made a commitment to them to lower our energy bill in data centers across the country. Now that we are able to do reduced footprint, dense footprint, more efficient storage. As one element, there's better storage servers as well. And now we're looking into some network elements doing the same thing, which is reduced footprint, reduced cost and reduced power consumption. We can get to those measurements faster. All right, Kareem Adil, it's been a pleasure meeting you. Thanks very much for all the time that you've spent with us. Okay everybody, this is theCUBE, this is Dave Vellante. Keep it right here. We're at Madison Avenue all day today and we'll be right back.