 Live from Las Vegas, it's theCUBE. Covering IBM Think 2018, brought to you by IBM. Welcome back to IBM Think 2018. My name is Dave Vellante and you're watching theCUBE, the leader in live tech coverage. Stay three of our wall-to-wall coverage of IBM Think. IBM took a number of conferences, Interconnect, World of Watson, EDGE, which was the infrastructure conference, brought them together. We're here to talk to Kayleen Sanchez, who's the vice president of IBM Enterprise Systems Storage. EDGE was your show. Welcome to the new world. Great. I know it's been exciting to be a part of the Think conference. Yeah. And what I think is great about it is we're talking solutions and the full stack, the full stack based on hardware, middleware applications, software, all of the feeders associated with delivering and users a solution. Well, I was talking to Ed Walsh earlier, actually yesterday, he came on. We weren't talking a lot of speeds and feeds, even though he's capable of it, but he's talking more about the adjacencies in IBM's businesses and cloud and artificial intelligence that are helping sort of uplift the storage business. I observed that having been an observer of the storage business for years, I've been hearing from big systems companies for decades that they're going to do that. They've had trouble succeeding, but it seems like it's finally taking hold. What's your perspective? I would agree. So the good comments associated with Ed is he's built a great team. We enjoy working together. He is fair, pragmatic in general. So we work to build collaboration within the IBM company to deliver solid solutions to end users. So he's done a great job. So what are the, you guys have reported four straight quarters of growth, not just like half a percent growth either, some high single-digit growth in some cases. What are the factors that are driving that? You mentioned sort of teamwork, culture, leadership. I'm sure there's some product stuff too from your perspective. Yes, so I actually, within my portfolio, I partner with Jeff Barber on, is like DS8000 enterprise storage and we see significant growth in the area based on our focus on flash and our investment with regards to flash optimization. The other aspect to really highlight is what we're doing in tape. And I know we've talked about tape before. Yeah, I know. Come on, let's talk about tape. All right, well, there's two components in that tactically, we're about to deliver a drive that's about the size of my hand that is called the LTO8, part of the LTO line, 12 terabytes for raw capacity. Yeah, so tape is interesting. I mean, the investments that used to be, back in the 80s, disk drive investments, all the VCs were pouring money into disk drives and heads and media, and a lot of those investments have dried up. You're not seeing the same types of investments. Tape, it's easier to do sort of funky things, multiple heads, drive super high bandwidths, do some sort of anticipatory indexing. Where do you see the use cases today? We've got blown out of backup. Where is it being used today? Is it archiving? Is it media? The NAB shows coming up, probably see a lot of tape there, but where are you seeing momentum for tape? So you are correct from a median entertainment perspective, NAB, that's a great industry we partner with. A few years back for LTFS, now Spectrum Archive, rebranded, a part of the Spectrum family. We won an Emmy. That's like... No kidding. I didn't know that. Yeah, we won an Emmy. So it's great in partnership with median entertainment. We're relevant there and our technology was relevant there. Now the other area for significant growth which helps feed those four quarters you referenced before is what we're doing with cloud service providers. We're relevant from a hardware infrastructure perspective based on tape. Tape is cool again. And there's a lot of companies worldwide who really believe that because it's all about big data storage for the right economic price as well as energy efficiency. Well, the gap between cost per bit for disk and tape is still enormous. It is. Tape is much, much, much cheaper and that's not going to change anytime soon, right? That is correct. It is much cheaper. So I'll give you an example. So basically less than a cent per gig per year. Now, I would actually even say it's less than a half cent. So it's just the economies of it. So a lot of what we do in talking about tape is the value from a cost perspective and the value you can provide a client where it's like, hey, they have big data, we can help serve it and we do that with tape. But is it, Kailin, is it the sleep at night factor? Like, okay, I'm going to put it in tape. Hopefully I never have to recover from it, but it's my media of last resort. I'm in compliance if I put it there. Or is that right? Or are people actually recovering from tape? Yes. Both. Yes, okay. So we were recovering from tape based on where fundamental tertiary storage for some of these enterprise clients where I have to discuss like tier management across primary, secondary and tertiary storage. So people think tape, classically, is an archive. Well, actually there's use cases that are fed by tape that kind of touch all the components of tier management. So I think it's more. It's more than just archive. It's big data. Now, let's talk about cloud. I thought cloud was going to take the on-prem business and wipe it out. What happened? It depends. That's what I like about IBM's perspective is hybrid. So we can serve both private as well as public clouds. And we also focus on optimizers. And what do I mean by an optimizer? For example, DS8000 in 2017, we delivered transparent cloud tiering, which allows you to basically take a primary device and treat every other storage component as a target to push data. Oh, by the way, you could push data to whatever cloud store in the sky that would be public or in some cases private based on security requirements associated with enterprise clients. So the criterion is largely security, not performance? Is that right or both? Both. It's a combination. It really depends on the use case that a client comes to bear or talks to us about. So I forget what you call it, but you guys had, early on, you had some automated capabilities, you had some magic heuristics to match data and device characteristics to put the right data and the right device. And you've extended that to cloud, is that right? So it's both policy based? So you are correct. So what you were referring to is easy tier management. So easy tier allowed you to move data to like a hotspot. Think of it as like a temperature rating. If it's hot data, it stays on flash or media types like that. If it's cold data, it goes off to ship off to cheap disc or possibly tape. Now our extension to that is transparent cloud tearing. I remember when you guys first announced easy tier, I'm thinking about it now. I talked to some customers and they said, yeah, you know, I want some knobs to be able to turn. I like to be able to manually move things around. And then this sort of machine intelligence wave comes through and people whose primary expertise was lung management or realizing that that's probably not the best career path for them. So have you seen customers become much more comfortable with that automation? Yes. There is an autopilot mode with regards to data management. But for some enterprise clients, I think I'm going to steal your word. They have to feel comfortable. They have to see that the right data was moved to the next tier and it's being managed appropriately. So some people like to like, for instance, your temperature reading in your house. Some people like that your dial is like 72.3, right? And you just know that temperature, right? Which is mental, right? Though, so clients were like that before, but with this idea of efficiency and we talked about flash efficiency based on our, one of our last interviews is that it gives you more time, more time to think about other things. And so EZ Tier provided us the capability, especially if you go autopilot, those end users can think about something different within their data centers to manage things differently, more efficiently. So it gives you time. And all I know is every Christmas, I pray to the Lord that I want 25 hours in a day. Yeah, so here, here. So the storage industry for years has been famous for doing more with less, constantly taking cost out. Guys who are whipping boys of customers and just squeezing every dime out of you as possible. You've made, IBM's made a lot of statements about Moore's Law, Moore's Law is waning, can't be as aggressive anymore, got to play different tricks. How has that applied to storage? How do you keep bringing costs out of storage? So I fundamentally believe everything old is new again. So we have to pay attention to the history of the legacy to really determine what the future roadmap is. And so what's nice that we partner with Ed Walsh on is talking about our building materials across our entire solution stack and ensuring we provide for exceptional efficiency. We definitely want, within IBM, to be the Toyota production system for storage. So it reminds me, say everything old is new or new is old. I remember a head of IBM storage one time who didn't know anything about storage and he admitted, I don't know anything about storage, he said, but I know this, it needs to be lightning fast, rock solid and dirt cheap. Has that changed and what's new in storage? So no, it has not changed, right? Though what we've been talking about is some really dirt cheap technology with regards to like tape, right? The last I checked, less than a cent per gig per year for storage management, that's huge, right? So that helps the wallet. But at the same time, there's some new future items, like we're wanting to play in the nanotechnology space, specifically to partnering with Sony, Sony Media, with regards to Sputter Media. So what people can go out and see when they have time is watch YouTube videos about what Sputter Media's about. Now, some of the deployment associated with Sputter Media was 220 terabytes for a single drive, that's our goal. So when clients come to us and say, hey, we want to serve or be served with data capabilities of like two X per year, well, we're at a point where we're going to blow their socks off because we're going to have an offering on the table tactically to be north of 220 terabytes per drive, pretty exceptional. What are some of the other kind of cool texts that we should be watching? I mean, you've seen advancements in file systems, obviously all the whole Hadoop and big data craze. Flash has completely changed not only storage but application development. You really couldn't be doing all this AI stuff without flash storage. NVMe, NVMe over fabric is coming in hot. You guys have done things like Cappy to get sort of atomic rights and capabilities like that. Again, game changing, geeky things that have business outcomes that completely change the application development paradigm. What should we be watching for from IBM? Some of the cool tech. So the other aspect that you've asked me in a prior conversation is about quantum computing. So we just need enough bits so they store those bits on us. So those are some of the early discussions about how IBM storage is going to play in quantum. Yeah, you got some cool demos here on quantum. It's kind of blow your mind demo, so check those out. Kaylee, I'll give you last word. IBM think, put a bumper sticker on it. So tape is not dead, it's sexy. And then also this other aspect of, I don't know, we can grow and so IBM storage is where it's at. And that's the reason why I remain here. Tape is sexy. Take it as tape is big and sexy. I know, big and sexy. Kaylee, thanks very much for coming back to theCUBE. It's great to see you again. Thank you, it's great to see you. All right, keep it right there, everybody. We'll be back after this short break.