 Hi, this is Dave Vellante, and this is theCUBE, where we extract the signal from the noise, bringing the best guests that we can find. The storage alchemist is in the house. Steve Keniston, who's with IBM. Steve, thanks for coming on. Thanks Dave, I've always liked being on theCUBE. Big day today, you know, we heard Steve Mills, we heard Ed Walsh, we heard, you know, from customers. You were doing a tweet chat. We did a great tweet chat today. So, give us the breakdown, what was the day like? A lot of very interesting comments coming through on the tweet chat. Things about like flash revolutionizing the world, what's going to happen with flash. A lot of the comments around the economics of flash coming up about what's going to happen to software licensing. A lot of the other thought leaders on the tweet chat talking about what's going to happen with the volume of flash that ends up getting sold now because of kind of what's going on in the flash industry. I really enjoyed this particular day because it really brought to light in a way that IBM is not necessarily known for doing, but really giving the industry a kick in the pants and saying, hey, wow, you know, this is really important and it's going to really revolutionize what's going on in the data center. So, you know, these tweet chats are great because they're open, right? They're like peer insights, right? Anybody can come in and I know that you guys were hosting it, but we had, I saw a guy from EMC on there, we had a bunch of analysts on there. Fusion IO was on there. Fusion IO was great to see those different perspectives and that's really what it's all about, right? It is. You're pushing your, you know, what into the stream. It's about getting the perspectives and debating and so IBM put forth today as a tipping point. What was the consensus in the tweet chat? Do people actually see flash as a tipping point? It was very interesting. I think that the notion of it being talked about as a tipping point became very obvious throughout the tweet chat. More and more folks were talking about, okay, then what does happen? How does that affect, you know, different parts of my infrastructure? If I do do that. Meaning it wasn't so much, I'll get there one day, it's about, hey, if I do do that, now what's gonna happen? How is this gonna evolve? So I think people started to embrace it and started to engage on it. Things about what happens to my data protection environment, what happens to my applications? How do I better write software? How do I become more competitive? All across the board. It was Oracle, like you said. It was EMC, it was Fusion IO, all having the same conversation. Obviously a lot of those guys wanting to get on board because flash is important to everybody, but a lot of the thought leaders and the analysts, too, talking about that as well. I wonder if we can compare what's going on now. I know we can, so let's do this. Compare what's going on now in flash with two, I'm gonna take two examples, Steve, that I know you are pretty familiar with. One is the whole data duplication trend, and then the other was the move toward virtualized storage systems. You know, people used to call tier 1.5. So three-par, compelent, left-hand, equal logic, XIV. Are there similarities, and what are they, and are there differences? Why don't we start with the deduplication piece? Yeah, so I think, and you and I have talked a little bit about this, especially because in the backup world, deduplication came on with such a strong force over a course of five years. So what we saw was in the industry, technologies and capabilities, and having worked at Avamar, I can speak to this, right, we saw a change in the industry from how data protection was going to be done to a really revolutionary way of doing it. We're going to do dedupe at the source and save you all these things, whether it be money, whether it be performance, whether it be storage space, and then what we saw was companies like Data Domain who came on board who then really thought about the overall process, right? Let's not disrupt the process. If I can unplug that tape drive and plug it into a dedupe appliance and get those same benefits, now all of a sudden the adoption started to pick up. Whereas, you know, after the three to five years, now all of a sudden, if you don't have dedupe in your environment, you kind of don't get it, right? It's becoming more and more predominant. But I see these technology swings become exactly like a pendulum. So whatever it was first, right, the pendulum was slow to swing that way. And then it was another capability, technology was like deduplication. Now it comes back a little bit faster. Then it was tier 1.0, and it's going even faster, right? And I see flash starting to gain more momentum in a much shorter period of time where folks are going to start to lean that direction. So the similarities there, you mentioned Data Domain, they crushed it because they were able to drop it right in without a disruption to your existing IT processes. Avamar did well, but never as well. And because essentially you had to look for disruption, had to be virtualization or it was essentially a rip and replace. So that's the similarity. The big difference that I see, and what if you comment on this, is backup really doesn't deliver a lot of business value and flash has the potential to deliver a lot more business value by breaking us, the stranglehold of what I always call the horrible storage stack. What are your comments there? So a couple of things. First of all, I totally agree with you, right? Name a company unless you're an MSP where backup makes you money. Name another place where spinning disk other than an MSP makes you money, right? It doesn't. But the value that flash brings, not that the flash drive itself makes your corporation money, but the fact that I can drive better performance, lower the latency, and again, re-architect my applications to actually have the ability to build in new features that drive competitive advantage, that's the real stepping stone, that's the point. And when you start to look at the economics and you're a small company, and that small company, take a bank for example, I choose the bank I do business with because they offer me a set of services that I can do seven by 24, anywhere in the world from any device that I own, right? And when the next bank wants to compete with that bank for services, they need to go to IT or their development organization and they need to ask for a set of services. That group better be able to turn those services on, lickety-split, if they can't, they lose competitive advantage. So now if I don't have to worry about architecting in performance and I have to think about all these things when building my applications, now all of a sudden I can focus on innovation. I can become more competitive. Now a small company, a small bank, can really come up with really cool new innovative features that can really outpace the big guy. So you're going to see a shift in the market where even smaller companies now become a lot more competitive. Yeah, this again is so fascinating to me because now, I want to go back to the other example I was using is the tier 1.5. You saw some huge exits. I mean, data domain was a big exit, two and a half billion, three par was right around there, the other one was just about a billion, some big numbers that we saw. XIV was actually lower, IBM is smarter in the acquisition front, doesn't like to spend as much. Now you're seeing companies, big companies, go out and buy in flash companies. You guys bought in Texas Memory Systems, you had EMC, bought in Extreme I.O. And not paying, I mean, I don't know what you paid for Texas Memory Systems, I haven't announced it, I guarantee it wasn't two and a half billion. So you're not seeing these big giant acquisitions yet, anyway, yet. What you're seeing, I think, is companies trying to get ahead of that. They're seeing the future and saying, no, why wait and shell out two and a half billion? Why buy it in now? So do you feel like the large companies are just getting smarter about that? Is it just that flash is so transformational or is the potential, it's not that the potential is not as great, we just agreed that it is as great. What do you think's going on there? Is it just that companies are trying to get ahead of the curve and not have to shell out two, three billion dollars? I think so, and I think you can actually almost equate this to like what EMC did with Avamar before they actually purchased data domain. So they knew they wanted to be in the data deduplication space, number one. Number two, they took a little bit of an innovative approach, right? So they said, hey, let's do it a little bit differently. They saw that that was fairly disruptive, right? And also created a little bit of competition inside their own company with other capabilities they had a la legato, right? So same thing with flash. Now all of a sudden you've got disk that can be tier one and you sell a lot of tier one, IBM sells a lot of tier one, but you wanna get into that space because you wanna be able to be on that innovative curve. And obviously you've heard a number of different times and a number of different reasons why IBM chose TMS. But again, back to the EMC example, I mean, yeah, EMC never didn't spend two and a half billion for Avamar, but they got into the space, they understood the space better and then once they really understood the space, that's when they doubled down and made that other acquisition. But I think you're seeing the same thing today. Yeah, and maybe it'll repeat itself. I mean, there's still, HP's really gonna use its own approach with three-par oracles, really not clear what they're gonna do. So who knows, maybe they make some blockbuster acquisitions. And you really, there's two big guys, right? There's Fusion IO and there's, I guess Violin with the two sort of out in front, but then there's a lot of other guys in there that have developing a lot of innovation. I almost feel like it's gonna be musical chairs. And when the music stops, somebody's not gonna have a seat. So anyway, congratulations, IBM's in early. And we certainly expect to see some organic development. We talked to a number of folks today about that. So it's a very exciting space. I mean, I'll give you the last word. No, it is an exciting space. And I'm looking forward to see again when I think about the pendulum and how this pendulum starts to transform IT. I talk about we started out in the mainframe days. We've gone all the way over to the right to the open systems mainframe, I call it. So now I can build on an open stack set of principles to be able to build out my infrastructure. And now, as we look to the future to things like software-defined infrastructure or software-defined environments, we're going to see new capabilities that Flash is gonna allow that SDE architecture become even more prolific. And I'm pretty excited about that. All right, the storage alchemist, Steve Kandastin. Always a pleasure. We go there again. Thanks, Dave. I really appreciate you coming on. Good to see you. My pleasure. As always. All right, everybody, thanks for watching. We'll be right back.