 Okay, we're back live here at Oracle Open World. This is theCUBE's exclusive coverage of Oracle Open World 2013. Oracle Team USA, Oracle Boat just won the America's Cup. We're here live in San Francisco covering all the action. I'm John Furrier with theCUBE, our flagship program. We'll go out to the events and extract a seat from the noise. I'm joined by my co-host Dave Vellante. Dave, they just won the America's Cup, came back from eight to one, tied it, pulled the ultimate comeback in international sports history. Tell us about the Red Sox beating the Yankees and then running the table, then winning the World Series. I mean, amazing, amazing comeback. Well, at least Larry's choice, not to show up for the keynote yesterday, I guess in a way, paid off. It would have looked even worse if they had lost by a mile today. But, you know, as we were talking about off-camera, it looked a little iffy at first, right? They had a draft in behind them, and I don't really know much about racing yachts, but... It was the first attack that made the difference. That first turn, they had the boat speed. The key was the boat speed, Dave. You know, looking at the race, the key was the boat speed, and the boat speed was really how what Oracle got to win the race. So that's exciting. Okay, and so we're back to tech here. Jake Cutrell is here. He is the member of the office of the CTO at VCE, and we're going to talk about Converged Infrastructure. We're going to talk about Oracle. Jake, welcome to theCUBE. Thanks, Dave. So what do you think of the race? Well, you know, when it comes down to eight million dollar carbon fiber vehicles flying at that speed and that velocity going down the line, it's amazing. I think of them like the Yankees of yacht racing. Yeah. Yankee? Well, we're not big Yankee. We're ready to talk, actually, about yacht racing. I think that's the analogy. Come back, thanks. Yankee fan? No, I'm just saying from a budgetary point of view. Yankee's numbers. Well, Dave, you know, last week we had EMC on early, and throughout the day we had Annie McCluron talking about the VCE. So you guys are out scouring the landscape in the office of the CTO. What's happening with the market? One, two. And then talk about the expertise in EMC right now. A lot of the database guys on board. What's happening? I think when you talk about the high-performance database side of things, you're going to find a few items that come to mind. First, of course, Oracle here at OpenWorld. So when we find that, Oracle here at OpenWorld, so when we focus on that as a workload of interest, going back to the early days of VBlock, one of those key workloads has always been Oracle. There's others, of course, but when we think about optimizing or tuning or making something that's a beneficial way to approach that application specifically, we've done that with general systems in the past, but we're starting to see from customers a demand for I've got to have something that is very Oracle-specific and aligned from a stack point of view to say I'm going to be putting my Oracle investments on that VBlock. Okay, so you got Oracle doing their engineered systems thing. Do they open you with open arms and so welcome you with open arms? Yeah, come on in, we'll help you integrate. What's that all like? I know Larry and Saf are good friends. We heard that yesterday. So clearly there's relationships at the Cisco, the EMC, and to some extent the VMware level, and what we're looking for really is just how do we accelerate the adoption of our customers taking on Oracle workloads specific to a VBlock investment. That's our tie-in as to how open arms it might be. I would say we're probably even putting with any other partner that might be out there. Yeah, because they want to make their stuff run faster and they realize it's not 100% it's not going to run on Oracle. Okay, so what does it mean to optimize for high performance database? Specifically, what did you have to do with Oracle to achieve that objective? Key things for us were really starting with what the database administrators' concerns were. They wanted to see certain things related to how they could load more effectively, faster, maybe in terms of how they would manage the landscape of the investment that's out there. So some of the things that we'll be doing, there'll be some stories about vision intelligent operation software, which you probably heard about previously, how that'll tie into things if you're looking at it from Oracle management perspective. But today, really the story is around getting those metrics that are important just to DBAs. And that's a big part of it. It's not the entire piece of it, but it's a big part of it. And we heard strong messaging actually yesterday on the keynotes from the executives at EMC about targeting the DBAs. Now, of course, you're in a similar boat. You've got the DBAs and the application heads really driving the decision making. So what matters specifically to a DBA and what are you doing in particular to address their requirements? Satisfying IOPS is probably the top item. Another one could be concerned with the real issue of the DBA. And I think that's one of the things that we're going to be doing in the future. We're going to be working on that. We're going to be working on that. One could be concerned with the relative spend of compute to those IOPS requirements that you're talking about. A good example came from a customer I talked with recently. They had a generalized system and they were able to do things like take the higher end blades and deprecate those for application development as opposed to the Oracle development because they were getting IOPS that were coming back in the array. They weren't necessarily thinking of it as a compute constraint issue, as a IOPS constraint issue. So by actually using lesser blades, they were able to satisfy the Oracle as a service side. Now, that was a neat thing for them, but how do we extend that to say, here's a purpose-built approach to doing and satisfying that one requirement. And so those are the things that are going into those specialized systems. So you specifically, Jay, mentioned IOPS. Yes. Where does latency fit in? Latency is a big piece of it, clearly. And so things that will be leveraging related to extreme IO, the SF side, that's the part we're picking up a lot of that. So with generally performance, there's a high level requirement on the DBA's list is what you're saying. Now, so this is a, the new product is called the 340, correct? So there's a 340 which represents the enhancements that have taken place in the VNX line, for example. And then there is the VBlock Specialized Systems for high performance databases. And if you look at the 340, that would be things that are based around the B series, the blades. And then when you look at the Specialized Systems, that'll be where we're including C series from Cisco, which are the servers. Okay, so the 340 is more general purpose. And then a customer would make the decision to go toward the Specialized System. A Specialized System would probably be one you would find where they have centered on database as a service, or they've got a database corner of a horizontal that all of the application teams would tie into, or they have a specific practice around database specific. And that would, again, go back to what you were saying before about the people that are driving those decisions, being specifically oriented to a concern for today, Oracle. But clearly we're going to be looking at the server long-term. So obviously those more intensified database environments and applications are what you're targeting. And they're high value, right? Driving revenue typically for companies. What kind of premium would I have to pay to get the Specialized System? I mean, even rough percentage terms. It's actually no different than if you were purchasing from Cisco or EMC or any of those parts and you put them together yourself. The difference is that what we're doing is the risk category. We're going to compare commodity to decommodatized. But so relative to other V-blocks you're saying you're not charging a premium for this. Okay. So it's pick your choice and there's no price penalty for doing that. Well, we also believe that there's going to be a question of what I can mix and match within that design. So for example you talked about the red stack view of the world where you really are staying between the rails from an Oracle point of view and this is a specific version of database, for example. We entertain having multiple versions of those databases on that same Specialized System. And indeed you could actually put applications or other things that would complement that on that same system. But you're adding, I'm going to ask you, you're adding value. Why not charge for it? Well, what we would like to see of course is greater adoption of V-block in general. We believe that mixed workloads are the way things are going to go. But we definitely see that there are those folks, those application owners, those DBAs that really believe that V-block Specialized or High Performance Databases is a product that they need to see. Jay, from a technology perspective, can you, an even ecosystem, can you compare and contrast the red stack approach from the VCE approach? So I would say that the largest jump would be if you're looking at, I want to do a little bit of VMware, maybe a little bit of bare metal, maybe a little bit of other, maybe I want to do some Oracle of a specific version, maybe I want to start using 12C, but I also have my existing investments in non-12C environments and I want to mix and match those when I'm ready, but I want to do it within the exact same physical container. So, deterministic power, weight, cooling, geometry, known spend to your point, whatever that margin might be and I do that and I execute that in a single experience with one company. That's VCE, so you're actually getting some variety of choice in what application workloads you can put to complement Oracle in that mix. Okay, so let me follow up on that. I'm going to go in and say, hey that's true, but we have end-to-end integration throughout the stack all the way up to the application with the only guys that can do that, we can take advantage of hybrid columnar and all this other cool stuff and we really are one company. Those guys, there's joint venture, V, there's C, there's E, how do you respond to that? I would concede that you are indeed one company. I would also say that we actually bring in best of breed solutions from a variety of places. I get to viciously decide what part of the investor stack I choose from. Which Cisco products, which EMC products, what networking technology from Cisco I bring into that and as I pull all that together in conjunction with Panduit, Corning and others I'm delivering a system that, thermally speaking, physically speaking is extremely deterministic and easy for you to consume right now. So Oracle is certainly trying to sell the world that it is best of breed. I haven't bought that yet. Right, I mean Oracle is best of breed at the compute layer, the storage layer. It's got work to do and arguably the networking piece, not even arguably, networking piece is certainly another discussion. Right, it's narrow. Generally the whole stack in terms of its use case is relatively narrow even though the applications within Oracle are broad. And then they come at that from a strength point of view. They're saying that because we're so focused I would call it myopic, but they're narrowed down and focused in that one area. What we see is that no customer that we've interacted with says they're just going to do that one thing. So Oracle spends a lot of money in R&D as you know. They're intensely focused and they're like EMC and we're in Cisco in a way. They're intense competitors, right? So what gives you confidence that you'll be able to maintain your best of breed edge over time? Talk about that a little bit. So we have access to 20 billion plus in R&D to kind of install of our investors. That technology when it makes the most sense for the workloads that we're concerned with for our customers. So today we're talking about Oracle, tomorrow we'll be talking about DB2 and beyond that we'll be talking about SQL Server. Maybe not even in that order, but we'll be talking about them. And we believe that kind of a mix and match metaphor is going to be something we're going to see more of, not less of. So I take, I know the EMC number. I presume, I don't know if VMware is in here and I'm sure Cisco is not in there. But EMC has 80,000 Oracle customers. So I guess you could, I'm sure there's a great partner. You can add in some numbers there for Cisco to add it to that number. So you have access to 80,000 customers. Oracle has 40,000 hardware customers. So you've got twice the access to twice the number of customers that Oracle has today. Now, those customers as we talked about earlier are largely infrastructure customers. But nonetheless, it's a big customer base. What are they telling you and what's the relationship with the DBAs and the application heads within those customers? So there's a good blog post that one of my protégés, Jeremiah Dooley, put out and it's got a little cartoon where someone defines a problem at the management layer they then go back out of the project team. Management team then goes back to the EA, explains that problem to them and the EA says, well actually that's not the problem. Then everyone assumes the EA is the idiot. So what we're hoping for is that we can simplify as much as possible that infrastructure. A lot of times in presentation infrastructure is boring. It doesn't mean it's not important and it definitely doesn't mean it's easy. But we think it can be really awesome. And so to do that we really have to not depart from a speeds and feeds discussion for say those are important metrics that DBAs are going to be concerned with. But we want to put them at ease. We want to make sure that application owners know that there's a way for them to also enjoy into that conversation and not feel that either one of those parties is necessarily coloring the discussion. And then the infrastructure should kind of blend away into the background. It's still important, it's there, but we need to make sure that the application owner can drive what the users are consuming. Jay, we had Steve Karamon before. He's the Oracle alchemist, I don't know if you know him on Twitter. And former DBA practitioner. We were asking him what DBAs care about. He said he agreed with you. Number one performance. Second thing he said is, the way he put it I thought was interesting. We at the DBA, speaking for the DBA community, we don't like bottleneck. We don't like that, we don't like being in that position. Proceed or otherwise actually. It's just not working for us. We've got to move on. So first of all, what do you think he means by that and what can you guys do to help folks like that DBA, that proxy for that DBA, not be the bottleneck? Well, we haven't talked about security at all. Another context for security might be how do I tenetize and carve up allocated infrastructure units more effectively so that I believe I can do more A-B testing. I think that's one of the areas that DBAs have been not necessarily stranded as far as resource concern. But to be able to say, I want to test an idea out. How can I effectively do that? How do I test for A and B, come back and do those tests in a rapid iterative fashion? Historically. And what we're after at VCE of course is to enable that type of consumptive behavior and obviously on the application stack side. Where I see it going is those will be improvements that take place over time because of specific infrastructure investments and I believe those investments will be V-Block systems. So where does cloud play into that whole discussion? The cell service animated discussion could go on for hours on that piece of it but I believe there need to be very quick tactical tools to allow for those DBAs to quickly get into their landscape, carve up the areas that represent those A-B tests, get back out again, but more importantly communicate that back up to the developers. So you think of what we do with release compatibility matrix is our tag release way of keeping all of these various pieces of software synchronization. I want to see that flow up and be part of the continuous delivery, continuous deployment side of things. So I wonder if we can sort of go back to that security discussion at the up level a little bit. You set off cameras as something that's near and dear to your heart so you've got some expertise there. Some of the writers in Silicon Angle were prepping for Oracle Open World said, you know, when you get a guest on that knows security we'd like you to sort of probe a little bit because it's an important topic and I never talk enough about it. So from your perspective, let's start with the state of security. What do you see as the state of security? What are the big sort of trends that you see there that the industry is grappling with and needs to respond to? Yeah, so I guess the elephant in the room is the fact that the Amazon booth is way down down that end of the hall and I still remember Living Social and I still remember that 40 hours later the Amazon team kind of floated out through the PR release mechanism. Oh, by the way, we have a cloud HSM offering, by the way, just so you know if someone had maybe used that maybe these things might not have happened and there's no way to back it into the future and say we'll do that differently next we've already done it, it's been exposed. What I think is going to have to happen is whether it's a posture, a hardening guide it has to be intrinsic to the manufactured experience. So I'm actually pleased that Oracle does have these manufactured system I'm pleased that IBM, that HP and others are doing it because it gives you a chance to put in some of those best practices and some of that enforcement of posture is day zero, not day zero from the factory it comes with at least some modicum of protection or some thought around what might happen once this goes, again, plug it in, log in and go. And if we know anything about the folly of defaults is that you can do things. And so... What I was going to say is I would like to see those defaults be sensible and be mindful of the current vectors and kind of the salvos that are being launched against infrastructures today. So if we can be party to that and be a part of some of these wider Cloud Security Alliance solutions I'd love to see more of that, not less of that. Partening the top, as John Huffman says. Part candy shell. The performance message has been pretty clear Extreme Software Cash has gotten a lot of buzz here. Going to a little quick explanation of what's going on with Extreme Software Cash relative to Oracle and then how does the V-Block and the VCE package move from this high end mainstream IT? What's that? Is orchestration the key issue? What's the core issue? That's great. One of the things that's going to be also part of a tag onto our last story that we told on the 17th of September was we announced our next Release of Vision Intelligence and Operations software which will include the beginnings of how we're going to allow you to do these kinds of upgrades in the field to take advantage of that A-B test again and say, hey, I see a feature. I might want to take that from EMC related to that Extreme Software. How can I bring that in and make sure that I'm not knocking down three boards and that's the part of VCE that's probably not as well-discussed but it's all that regression testing we're doing behind the scenes, all that pre-engineering and validation on their behalf. So when new things come out, we'd like for our customers to be able to take advantage of it immediately. Dave and I always talk about VCE because we've been following since the Cube started in 2010 at EMC World and went to Sapphire. We talked with Tom Packett, Levi's at the time and VCE really, Vision was great but a lot's changed in four years so what's your take in the final two minutes? Let's talk about kind of what's changed and how has VCE evolved at a high level? What's the positioning technically? What's going on under the hood? Obviously a lot's happened with VMWare now, Pivotal. What's the head of it morph? What's it morphing into? Thank you for the Pivotal segue. So let's just do it with some perspective historically. 2010 I joined the company. The biggest drive I could source back then about two terabytes, big drive. Now we're going to be seeing things like a two terabyte blade that'll be ram. Within a couple more years I fully expect those new developers using things like these Pivotal stacks that are going to demand a two terabyte heat. Not ramp, heat. Just a heat for one application and one application because they want to do that type of in-memory manipulation. They're going to want to crawl across the wire, RDMA that stuff. See it, pull it, manipulate it, shove it back in somewhere else. So I see that progression taking place and each again, our investor technologies is aligning in those directions. So for me it's exciting to kind of see how we can put together in our timelines customer facing road maps and then what we also keep that we know about on a three to five year basis. And I know that over time it gets really complex to explain it all but our goal again is to package it for easy consumption. Yeah I'm at a high level. Dave and I try to compartmentalize it for the average IT folk out there and that is that hyperscale has really laid out a great vision for scale out. You see Facebook, these guys building their own stuff. Absolutely. If you're an IT enterprise I want hyperscale but I don't want to do all that work. True. That's ultimately what they want. It's a packaging discussion. I mean if we look at where things are flowing towards, there's a commodity curve and a decomodatized curve. And I think that over time you're going to see things like open hardware, open rack, open data center initiatives. There are going to be things that fall out of that. There's an acronym SUPE all around it. Everyone's kind of vying for the opportunity to be a party of that. There's legacy too. I've got red hat over here, I've got some storage and mix and match but at the end of the day, talk about developers. They need a scalable infrastructure under the hood from a compute standpoint storage and networking so they can do whatever they want. They don't need to do Python, Rails, whatever they're going to do. They need to do that. That's kind of where it's action. VC always had that nice package but again high-end. Ask a 20-year-old Ruby developer if you even know what a LUN is. No, they don't. Exactly. That's why DevOps is so hot right now. They want WIM based infrastructure. I desire it. I want to go get it. Today it's a credit card transaction for a lot of places and I think what's going to happen in the enterprise is they're going to be hiring millennials and increasing numbers. They're going to have to satisfy that demand. Since the Nassir acquisition, you saw Software Defined become the big thing. Final segment here, real quick rip on Kassi. Nassir created that boom of awareness of hey, you know what Software Defined and Virtualization being a key enabler in that and Martin Kassada said on the Cube at VMware last time we were here, hey, Virtualization is still the enabling technology. So Virtualization is not going away. Where is VC going to go with all this happening with Virtualization with data? The heap thing you mentioned is a great example of how coders can leverage resource. What's the big challenge? Is it orchestration? Is it still pushing Virtualization to the end? What's your take on all that? So I absolutely believe it's orchestration oriented as the challenge. Even if you look at Gartner's numbers when they were grading the different types of cloud management platforms that are out there, nobody gets 100%. Nobody. You know, I think at best 70% coverage from VMware today in their most gracious allocation of where they do their definition of what a management platform would look like at maturity level. So whether it's a CA a BNC, Cisco's UCS director, all the investments that have been made by EMC and their, you know, their mergers acquisitions, someone's going to have to put together something that's going to allow someone to carve up this stuff and make it easier to consume and use. And that's where I think the, that's the, that's the most awesome part of the market right now. Well John, we said several years ago that that whole management space was jump ball. Yeah. Yeah. So let's talk about just final question before we break here. EMC's always had a great office of the CTO since I've known EMC over the past decade and recently, past five years scouring the landscape. Yes. Right. And Joe Jucci's you know, hardcore about making sure that they're reinventing the future and investing. No stone unturned. We heard Annie earlier on Annie McClure talking about some of the database chops are bringing in really large scale, not kind of like Johnny come lately's in the market. Really deep expertise and it's all about EMC ventures, right? Sure. The office of the CTO had a tight relationship, always had a tight relationship with EMC ventures. What's your relationship now with the EMC ventures? How do you talk to those guys? Are you involved in some of the investment discussions? Can you share a little bit? Were you reading my inbox this morning? So yes, you know, clearly it's going to come up. Cisco has an arm. What was your email first? Joe Jucci said do some investments. No, no, it was, it was it was related but unrelated, I would say. And no confirmation or denial of that. But what I would also say is, you know, Cisco has an arm, Intel has an arm, VMware has an arm, EMC has an arm. What we're focused on again is when we come across things we have a lot of customers we interact with, names come up and companies come up that have really just beautiful elegant solutions that may have occurred to someone at some point in time but if they've really got that vision, they've got that thought and more importantly they've got some existing customers they've already talked to, you're going to see a lot of those kinds of introductions taking place. So I would consider us to be part of that canvassing operation. We get to talk to a lot of people, see a lot of things. Final question, what fruit is going to come off the VCE tree in the next couple of years? Also you have a great investment, you've learned a lot over the past four years, you have great big deployments, some of the numbers are significant, the dollars is good, Joe Jucci is going to brag about it on stage. You know, a lot of Christmas early on but you know, it's kind of in the right to be directly correct as with the trend line. So what's the fruit going to come off the tree in terms of products, value, dollars, you know, feel free to share. I think if anything, if there's an assumption up front that how you consume a unit of IT in a data center is footnoted as comma a V block, I think we've done our job and that's one of the jobs, there's many more to come but simplifying that data center allocation unit, that's a big part of it for us. Virtualization is still a driving trend, this is the cube obviously we know V block, virtualization is going to be key, virtualizing storage, virtualizing databases, virtualizing everything up to the stack, giving the developers the freedom to take advantage of an amazing amount of RAM, persistent flash and ultimately still spinning disk and then as Jeremy Burt says, maybe some tape later on thrown in, but tape sucks as he says, but J, thanks for coming inside the cube, really appreciate it. EMC is doing great stuff, again, EMC great partner of ours for supporting the cube, we want to thank you guys and DC in particular, it's been a great run, we're going to continue to the cube next year as well. This is Oracle Local World, we're going to be back with more live coverage day three here in live in San Francisco, fresh off the America's Cup ultimate comeback win, San Francisco's on fire, big party tonight, sure Larry else is going to make an appearance at some point. This is the cube, we'll be right back with our next guest after this short break.