 Live from Austin, Texas, extracting the signal from the noise. It's theCUBE, covering Dell World 2015. Brought to you by Dell. Now your hosts, Dave Vellante and Stu Miniman. Welcome back to theCUBE. I'm Stu Miniman here with Dave Vellante. Happy to have back a long time CUBE alum Travis Vigil, Executive Director of Dell Storage. Travis, welcome back to the program. Thank you, thank you for having me. All right, so Travis, you know, not much new going on. I mean, you know, Dell, of course, you know, we've talked to you for years about Dell Storage. First of all, congratulations. Not only is Dell, you know, the number one in kind of, we say overall storage with combined capacity shift. But, you know, now if we look once everything plays out, I mean, you guys are like the biggest player and probably the largest share in storage that we've ever seen since external storage has been announced. If you look at what happens with Dell and EMC. So, can you just, you know, where's your mindset at today? I'm just extremely excited about the opportunity going forward. I mean, I think if you look at the proposed transaction, the amount of new spaces, the combined activity can address jointly is just enormous. And so, you know, I feel really good about where we're at with our storage portfolio today. We had a couple of big announcements at this show that I'd like to spend some time talking with you guys about. And, you know, I think the future just gets brighter, you know, as you go forward. Yeah, absolutely. So, yeah, I mean, let's talk about the announcements. I mean, you know, some of the big spaces, if you talk about kind of the growing trends in storage, two big ones, all flash arrays, and kind of the converge and hyper-converge, and you had announcements in both of them. And before we do that, can we just, for the audience, so you had sort of equal logic, compelling, you bought two companies, you've brought those two architectures together. We have been, yeah, that's a good point. You've really, you know, reorganized and sort of rationalized the whole development, you know, team. You still got a small team in Nashville, New Hampshire, you got a big team in, I think, India, Bangalore, India. So you guys have been on that track for a while now, with a single architecture. Can I say that? Is that an active way to describe it? Yeah, that's how we've been talking to customers about it for the last, really last two years. We kind of came publicly out and started talking about it, you know, in the public, although we had been talking, of course, with customers under NDA for some time before that, about 18 months ago. And the journey we've been on is, if you look two, three years ago, we had three different architectures, and we said, from a customer's perspective, it would be a lot better if you could have value investing in one that transferred to another. And so we made the decision, hey, we're going to go to a single architecture, we're going to put all of our engineering resources and focus on bringing value to that architecture. We're really going to get ahead of the curve on flash. And we've been going down that path for the last two years. And if you look at the portfolio today, what we have with the SE architecture, SEV2000, SE4020, SE8000, and the recently announced SE9000, really spanning from $10,000 selling price all the way up to $100, $150, $200,000 selling price. With a single architecture, a customer can invest in and have data mobility, common management across that single architecture. At the same time, what we've been doing for people that were focused on the PS series architecture, which is a lot of our installed base, we have made it very easy for them to come forward to the new Dell storage architecture, which is based on the SE architecture. Providing, just with this announcement actually, we provided the ability to thinly import a volume or a loan from PS series into SE architecture, very simple operation for a PS series customer. And I've also announced that early next year we'll provide both common day-to-day management, single pane management for both PS and SE series, and bi-directional replication across the two platforms. So what you've seen is we have simplified the portfolio, but at the same time provided a path forward that allows a customer to realize a long-term ROI when investing in Dell storage. Because we know when a customer invests in storage, it's a five-year, six-year, seven-year time horizon that they're looking at. So we're very cognizant of that, and I think we're doing a really good job of moving customers to the Dell storage architecture. So Travis, there's a few transitions going on right now. We look down at the architecture and we've been talking about disk-first flash a lot and some of the all-flash architecture. Then you look at something like Hyperconverge, we position added, I mean, that's the death of the storage array as we knew it, going to more of a systems buy, of course, leveraging compute, which you guys have great position in. Where in the conversation when you're having with customers, how do they think of storage? How do they buy storage? How do they want to buy storage? You know, Hyperconverge is no longer, okay, buy a box, how many years migrated? It's more of creating a pool of architecture, adding, removing, and upgrading becomes a little bit simpler than we have before. So I'm just curious, what you're seeing out in the customer field, of course, there's a lot of changing dynamics, but what does the storage buy and sell look like in your customer base? So I think if you kind of go up a level, it's been the case that customers don't really buy storage. They do new deployments, they deploy new applications that happen to use storage. And more and more, that's becoming the conversation. How does this infrastructure support my workload? Does it give me good TCO? Does it give me good performance? Does it give me good dollars per gig, both now and in the future? And I think you're exactly right. The two major trends that we're seeing at the highest level is that there's a trend towards, what we refer to as a new architecture, which is largely software-defined, Hyperconverge like we have with our XC series, a web-scale converged appliance is a good example of that. And then you have the current architecture, which is more best-of-read server storage and networking. So when you ask me what customers want to talk about it, a lot of the talk is around software-defined storage, a lot of the talk is around Hyperconverge, but we have an investment in the current way of doing things that's been built up over the last 10, 20, 30 years. People are going to continue to build on, add on, deploy infrastructure that way, because it still does have value, especially if you need to scale up to large amounts of storage in a single array. So we see it not an either or, but a both and, and it will be a both and for many years to come. So I think the announcement this week really illustrates what we're doing on both sides of that equation. If you look at what we've done with the SC9000 announcement, 40% IOP improvement from the previous generation, we had announced earlier this year that we have the lowest dollars per gig for Flash, even when we compare our dollars per gig for Flash RAW versus competitors, allowing them to use their deduced rates. We bettered that with this release because we had the first release of active data compression on the SC series. And then on the more software defined Hyperconverge side, we announced two new models for our XC series. One is the densest model that we've had today. It's a 2U4 node, what we refer to as the XC6320 model. Any deployment that has rack space requirements, whether that be a large enterprise, a service provider, even a remote office branch office that really needs a very dense form factor to be interested in that. And we announced two all Flash variants as well, one based on the XC6320 and one based on the XC630. So you see Flash on both sides of the equation, kind of the architectural decision at the top. And we're going to continue to innovate, support and enhance on both sides of that equation. So let's talk about Flash, let's talk about Flash. So we had Mark Lewis on yesterday and he was saying, listen, Flash is not disruptive, right? It's existing guys, they all get their Flash strategies they're going to do very well, which was kind of an interesting comment. It does change everything. So we want to talk about that a little bit. Your Flash strategy leverages the tiering architecture that you guys have put forth. You got SLC and MLC, you differentiate those. What are you seeing in the marketplace? That's how you drive your cost down, you're adding data reduction as well. That's exactly right. What are you seeing as far as how customers are actually using that? Is that sort of strategy, the right one? Are people actually, you're seeing right intensity, requiring that type of split. What are you seeing in terms of reliability? We're hearing that actually Flash is going to be much, much more reliable than the spinning disc ever was. Give us the update on Flash. Yeah, I mean the Flash transition, when you look at the current, like an array, like the SC9000 has been very fast. Frankly, probably a little faster than we anticipated. We've been on the forefront of that transition. You mentioned, if you go back two years ago, we were the first vendor to be able to use both SLC and MLC drives in the same array. We have a very nice tiering and virtualized architecture in the SC series that allows us to steer rights to more right-endurant media and reads to more read-endurant media. And so that's what we did initially with SLC and MLC. As we've seen the market ramp, in early days, there was a premium. You really had to go after those workloads where more transactions per minute equated to better bottom line, right? And so that was kind of the early adoption of Flash. What we've done with our strategy is, there's a lot of apps out there that would like to be accelerated, but really, 15K may be good enough. So how do you drive the cost of Flashdown so that it's lower than 15K RPM drive? And that's what we've been doing. So you look initially SLC, MLC, and then we found, via our phone home data with our customer base, that there were a lot of customers out there whose right intensity could be serviced by a single tier of MLC drives. So just, I guess it was just about a year ago, we came out with a $25,000 starting price, all Flash, MLC array based on our SE4020. It's been a very good seller for us. And we've seen an enormous increase in Flash Attach over the last year. If you look at the revenue associated with SE series Flash arrays, it's grown 80% in the last year. And this was pretty TLC, right? I'll talk about TLC in a second. And so I chalked that up to the fact that we were innovating, bringing down the cost of Flash, making it such that Flash was really, at worst, as expensive as 15K drives. And in many cases, less expensive than 15K drives. And then we just were able to bring that lead even further with our recent announcement of TLC drives. Again, that highly virtualized architecture allows us to use right-intensive media, read-intensive media. So I anticipate that 15K drives are going the way of the Dodo bird and 10K drives are not too far behind. So maybe within 18 months to two years and then 7,500 within four years. Yeah, you know, when it gets out of a couple years. We predicted 15K crossover would happen last year. Yeah, and I think it did. We predicted that, by the way, in 2009. Everybody said, you guys are nuts, right? I mean, like going on the wiki, check it out. So we've got our latest Flash, you know, disruption forecast out there. Yeah. Use the disruption word. It is disruptive to hard disk, that's for sure. Yeah, I agree. And I think the predominant deployment scenario we're seeing today and it will continue into the future is a combination of Flash and 7.2K RPM drive. If you look at our Flash attached today, 85% of the time when we sell Flash, it goes out with hard drives attached and the vast majority, I would say, close to 100% of the time, those hard drives are 7.2K. Yeah, and the hard drive guys are doing some gymnastics to keep the 7,200 stuff alive and probably. I mean, you just saw WD just paid $19 billion for sandage, so there's definitely change going on in the market. It's an exciting time in the market. Travis. Travis, you have a hard stop, I think. I do, yes. So we have to wrap, so let me just, let's let them go. So thank you, Stu, thank you, Travis, for coming in. Keep right there, everybody. We'll be back with our next guest right after this. This is theCUBE, we're live from Dell World 2015. Thank you. Thanks.