 Hi everybody, this is Dave Vellante and we're live at VMworld 2011. I'm with wikibond.org and I am here with my new co-host, Steve Keniston. Hey Dave. Steve, I feel like we're back doing running data. No kidding, all right. Thanks for coming on and setting this up. So we have these spotlights this year, as you know Steve, and spotlights are designed to be in-depth segments to really go deep, help customers understand a particular topic, how they can most take advantage of it, what some of the use cases are, what some of the proof points are. These are sponsored segments. IBM is sponsoring the storage optimization segment. And so basically the way that we set these up is we have an initial overview of the marketplace and then two segments with subject matter experts and then we have an independent panel on there. So thanks for coming on, I appreciate it. My pleasure. So let's start with sort of what's going on in the marketplace and we'll just sort of banter back and forth. So on this slide I just put up some of the big picture trends. You've got cloud, you've got big data and they're really accelerating the data tsunami. I mean everybody's seen the IDC data, the up-and-to-the-right charts, which I know you love. Hate the up-and-to-the-right charts, but it is true, right? People are trying to figure out how do I get more data online so that I have access to it and that accessibility is slowly moving from the backup arena to just to be able to get to it to be able to, now how can I analyze it? Which I think is what really spawned this whole big data movement. Yeah, now the second point I have on here is that optimization is obviously going to lower cost. And now a lot of people think that, well, you lower cost, you cut the cost per gigabyte. A lot of the sales mentality is I'm going to sell less gigabytes. Yeah, and I think we've proven time and time again that that just doesn't happen, right? What happens? When we introduced deduplication, obviously when I was at Abamar, right? Now all of a sudden the backup world started to grow because now not only could I deduplicate it and put it on primary disk, I now had an opportunity to replicate it and because I wanted it online, I could efficiently get it to another location. I effectively created my disaster recovery plan and I had the data in another location. Same thing happened in WAN optimization. Same thing happened. I could now put compressed data on tape. Let's put more data on tape, right? So, optimization is becoming table stakes is the point on this slide, but it continues to be hampered by overheads. You have a number of companies who are, including optimization capabilities, let's say dedupe or compression in the system, which is, I've always said, where it belongs, but customers have to be careful. There's always a but, try it, see how it performs. And so I know one of the areas that you have emphasized is performance and we're going to talk about that. Yeah, I think there's always a trade-off between, when computer science, whenever you do anything, there's a trade-off between optimization, whether it be chip optimization, storage optimization, and then on the other side, efficiency, right? How fast or how quickly can I get it done? I think the key thing to always remember, and I think it's one of the core values of when we were developing the real-time compression technology was, you need to always make sure that customers buy storage for two very key reasons, performance and availability, right? And you got to make sure that you maintain those things because customers don't want to see anything different. So, this slide here that we have talks about how organizations are drowning in data. I like it better than the up and to the right slide. It's a conceptual slide. But it basically says, okay, computing power is going crazy. You've got data going like nuts. But the amount of data that's actually filtered, that algorithms allow us to actually make sense out of it, is limited. And that gap between the amount of filtered data and the amount of data that we have is this information gap that causes problems for organizations, right? What happens there? I think, ultimately, right, what you want to try to do is close the gap. But because we talked about data is growing so fast and the tools that we have to be able to access that information and understand that information is more and more difficult to be able to do that. So, this slide actually really talks to the fact that the more data you have and the ability that you're actually able to have to be able to make more mistakes, making a mistake, I always even said this in business. I don't mind failing at a company, right? As long as I learn from that failure. If I can learn from the information that I have and make that failure, I now know something about the data that is untrue, which helps me know that some other thing must be true, right? So, it's the importance of being able to access all of that content and that gap there, the tools that are available to be able to get to the information. However, it's stored, it really becomes the key. And that leads to data value. The more context you have around information, the more valuable it is. I mean, it's really a simple concept, but why is it so hard for organizations to actually implement that concept? I think you, I mean, we talked a little bit before about budgets being flat within storage organizations, right? I don't have all the money that I used to be able to spend to keep everything online that I would want to keep online. And then I coupled that with the fact that drive capacities and the fact that I can't store more in that particular footprint. And coupled that with the fact that I'm not going to be going through, again, the economic time keeping my people down or whatnot, I can't go forward and continue to add more resources to be able to do more work. So I'm kind of stuck in this limbo where I have these set of processes that allow me to store a certain amount of capacity locally. And I kind of grow with that as I see fit of how my data just normally grows, which doesn't help me get ahead of the curve. I just try to maintain pace. And then at the end of the day, I'm trying to protect that and back it up and do all those other ancillary things that are required with maintaining a storage environment. I never seem to get ahead of it. Now I talked earlier about the overheads involved. And we've done an extensive research on Wikibon. We have a methodology called core capacity optimization, ratio efficiency, it's a measurement, essentially an ROI measurement, but it takes into account the cost of overhead, the cycles, the CPU cycles that you have to use. And so we live in a world of real time, real time decision making, the real time web, everybody wants real time. And so what we're seeing here is a real push toward getting those overheads down and having performance really be transparent to the users, aren't we? Absolutely, you can't stress enough this notion of having a real time platform. Being able to have access to information is one thing, but being able to have access to all of the information that you want in real time is just tremendous. If you go back up a couple of slides and you take a look at that quote, every millisecond gained in programming trading, applications is worth $100 million a year. So back in the beginning we were talking about what is the value of having that information available online to the user? Every piece of data means money. So how do you make these vendor, these companies, more business efficient, that's what they want. I can't spend as much money as I used to, how do I spend it focused so that I'm more business efficient? And you got to make technologies that drive that business efficiency. So let's lay out the landscape here. We're doing a panel, I have a panel at SNW, you're on it, Craig Nunez is on it, Jared Floyd from Permabit is on it, and Larry Freeman from NetApp is on it. So let's kind of lay out the landscape. Let's start with NetApp. NetApp's got compression, they got Ddupe, it's built in, you got Albirio, which has basically an SDK, which it's shipping, IBM real time compression. You got, I think EMC's doing some stuff, right? With their compression and their duplication, correct? Built into the systems, Dell acquired Ocarina, which is more of a... Appliance based today, but Ddupe, an intelligent Ddupe that has file aware. Really for archive data, so that's really not in that real, of those that I've mentioned, the guys that are real time are really Albirio and IBM real time compression, right? Those are the two. Albirio's Ddupe, IBM's obviously compression. Yeah, so that's kind of the landscape. We've said on Wikibon, I don't know if you can comment, but we've said that eventually these things are going to be embedded into systems, that's really the direction that this is going. It really makes a lot of sense, make that transparent to the users. So some of the customer issues that we're looking at, the recession is still fresh in people's minds. I mean, and if you can optimize the storage, you can save some money. I think every customer's going to start asking, well, why isn't all my storage optimized? Are they asking you that? They are. And we've got macroeconomics uncertainties. We had Tom Georgins on today from NetApp. He was talking about the macro trends. There's still a lot of uncertainty. I'm not saying there's going to be a double dip recession, but few people are worried about that. At the same time, data growth doesn't stop, right? Even in the recession of 2008 and 2009, data growth continued at 50, 60, 70%. So the need for optimization is greater than ever before, and as we talked about, the world wants real time, the world needs real time. So these are some of the things that we're going to explore in this spotlight. Anything else you'd add, closing thoughts? No, I think you've nailed it all, right? It's not just even about, you know, you talk about data growth. It's having that data online that customers want, right? Given the economics, right, how do I make my business more effective? Now if I can make my business more effective by getting more out of my data, then I want to be able to do that. Well, how can I do that if I can't actually go and add to my capacity? If I don't have the power, the cooling, the floor space, whatever it is, if I can't do that, how do I end up doing more analytics on more data? The answer is optimization. Somehow you have to optimize. All right, this is Dave Vellante from Wikibon.org. I'm here with my colleague and co-host, Steve Keniston. Steve, it was great having you back. Let's do more of these. We got a couple more sessions. Keep it right here. We'll be back with the in-depth spotlight on storage optimization.