 So we'll just give one second here. So hi, I'm Steve Herrod. I'm the CTO at VMware. I've been at VMware. I was involved with the starting of it at Stanford, and then I've been there for about the last 11 years. And it's a pleasure to be here. I was just telling some folks this event has a similar feel to the early days of VMware, which is one of the big conferences we hold. In fact, it was in San Diego. We expected, I think, 800 people and 1,400 people signed up. So it's very similar on a bunch of fronts. So anyway, this is probably a lot of people are wondering what VMware is up to around OpenStack. We've had a lot of traffic at the booth, and my goal in this next 35 minutes or so was to share technically what our strategy is, and then also where we're working with the community today and where we'd like to going forward. And I'm glad there's a lot of interest in that as well. So I'm just going to quickly go through who we are, why we're here, specific technology areas that we're working on, and then some ideas to work together on next steps. And I think we'll have about five minutes for Q&A if anyone wants to ask something on that front. So maybe I could just do a quick level setting. How many of you here are using VMware in some capacity or know about VMware? OK, super. That's great to see. So I won't go through the full thing. But I'll give you a little insight into where we come from, because I do think that really sets a stage for what we're wanting to do as we go forward and a little bit on where the strategy is changing. So for those of you who don't know, we were started in 1998 in the Bay Area out of Stanford. And the code name for our project, I always like this because we did not invent virtualization. Any of you that have been around for a while know that it has been around for a long time. So we called it Disco, because it was a cool idea from the 60s and 70s that we wanted to rekindle and make cool again. So that's why it was there. And the way we started was really thinking that we could ultimately be in the data center just as mainframe virtualization had been. But we really started with developers and running Linux on top of Windows. So that was our first slogan of the company. I kind of fast forward through what's been a very wild ride. Today we're about 13,000 employees. We have 4,000 developers and engineers at the company. And what a lot of people don't know is that we have over 40 different products spending a lot of different areas. And that's very relevant, because a lot of them are things that I think would help the OpenStack community quite a bit. So I'm going to give you a little flavor for the areas where we're focused. A lot of you might only know us for one product or the other. And then lastly, I just want to call out, our whole life since we started has been around partnering. It is really core to the way we run things. And again, these are stats a lot of people don't know. 85% of all business that VMware does is actually through partners. Whether it be through server or storage vendors who are selling software. It could be through value-added resellers. It could be through different groups like that. So 85% of all things we do, revenue-wise, goes through partnerships. Now the second thing is that we've had to from day one be very, very open around the core hypervisor work that we've done. A lot of you know it by ESX. It's part of a broader virtualization solution called vSphereNow. We have more than 20 different SDKs that are in place there across everything from writing device drivers to integrating with power management systems to monitoring and everything in the middle. We've also had to work quite a bit on doing hardware compatibility lists that have literally more than 2,000 entries on it from different storage arrays to different servers. And we have a massive lab that focuses on interoperability, which is something that obviously the OpenStack community is working on and very key to giving guarantees to customers. And then lastly, the thing I want to make very clear too is that a huge part of our company is focused on open source. And you might not know that a lot of these are driven by VMware. And I'll go through them in a little bit more detail. But whether it's through Spring Source, which drives Tomcat in many ways, Rabbit and Messaging Bus that many of you use, a lot of contributions to Linux to make it run well on top of vSphere, Cloud Foundry I want to spend a few minutes on, Redis, et cetera. So we are not strictly a closed source company. We're not strictly an open source company. We're a blend of both. And we want to participate in both in different ways. So that's just a quick backdrop on who we are and what's going on here. From an overall company standpoint, we're really excited to be really an enterprise-focused company. That's what we do. We don't do consumer products at all. And we've really been looking at the three big trends that are hitting everybody right now. And I have to say, after 25 years working in the overall infrastructure business, now is the most exciting thing, I think, in a long time in terms of all the disruption happening. So we focus on three different areas. The core of the business really is around the transformation of legacy data centers into new things. That might be first virtualizing them so that they use their hardware more effectively. And then, obviously, the whole talk here has been, how do you then make it a more agile IT department that can respond to demands automatically as you create a private cloud model within existing data centers? Or it might be working with partners on public cloud offerings. And I'll show you a little bit more detail on that as well. But the two areas you might know a little bit less about us on, we have a very substantial business looking at how you rewrite applications as we head forward in the future. And everyone who's in infrastructure, it's too easy to forget that infrastructure is simply a means to an end. It is a way ultimately to run applications, existing ones or the class of new ones. And the new applications, I think, as everyone knows, have a variety of demands on them. They have to look good on mobile screens. They have to combine a bunch of different services together. A lot of times they're doing interesting analytics with big data. So a huge focus of the company now has been on what are the next generation of applications going to look like in investing in those tools and those capabilities. And then lastly is something that hits us all more personally, which is how do I ultimately access machines, applications, and work with things? So we have quite a bit of things going on there. So just to map a few product names if you're aware of what we do to these areas. Again, the core focus of the business has been around vSphere, which is the way that we provide virtualization. We have a vCloud Suite, which is something I'll talk about in a moment, a variety of management and automation tools that work on top of that as well. In the space of how we're modernizing applications, our focus is on, as I said, the spring framework, a way of writing Java in a very efficient way, we believe. RabbitMQ, Hyperic, if you've heard of that before, and Cloud Foundry, which is something I want to spend some time on. And then lastly, a lot of you probably have used Workstation, Fusion, or Player, some of these tools that are for individual productivity. Very much focused on developers getting their job done. But also a lot of what we do also is how do you remotely access enterprise desktops, enterprise applications, working on mobile phones as well. So this is the breadth of where we work. Kind of put in your head, probably half of our engineering is on the bottom level, and then about another quarter on each of the two levels above that kind of give you a flavor for things. So that's the focus here is. Now I want to talk about vSphere and where we are in the enterprise at the moment, just give you a little flavor for why we've been excited over the last 12 years on this. So we're used by kind of officially 400,000 different companies, which means, by definition, really a big span across geographies, sizes, verticals. And I got to say, as a software engineer, being able to work in something where a small company focusing on oil exploration or a large company doing retail gets to use it, that notion of a horizontal platform is exciting. That's obviously something that I think OpenStack Community is striving for as well. It means that we get to, we're running the manufacturing lines at LEGO. We're used for Ducati cycles as they go and create their storefronts. I'm a Dallas Cowboy fans. The whole stadium uses VMware to virtualize things. And it really goes across the board. It's even being used in ATM machines today. And that's because there's a core efficiency as you go from abstracting hardware out and being able to use it more efficiently. Now, the other thing that's going on, we are used quite a bit in public services. And I'm careful not to call this public clouds because these span a whole variety of different solutions. These are system integrators that do outsourcing. They're traditional hosters doing co-location services with virtualization, all the way up to people who are offering pure private clouds, public clouds, excuse me. And the point here is that there are a huge set of models around public cloud offerings. A lot of them, all of these that I'm referring to here are using vSphere as the foundation, but they've built all sorts of different scaffolding on top of it. Most of them have their own portals, their own automation systems, their own different things that are on there. We have a product that's called vCloud, which a lot of them are using now to automate a lot of these pieces. And obviously fitting into an open stack world is something where they could use that as an option as well. So we are squarely in this space from a virtualization platform and are excited about extending it and making it even more efficient on that front. And then lastly, just again, I've been at the company for a long time, so I get very humbled when I think about just how pervasive these things are now. So if you talk to the analysts right now on the server side of the world, somewhere around 60% of all applications everywhere are now running virtualized, which is hard to believe. Somewhere between 50 and 60% if you look at all the different ranges. But that means more applications than not don't talk to the underlying hardware. They're talking to an operating system that in turn is talking to some sort of virtualization layer. So that just really opens the door for dramatically rethinking how you write apps, how you run them, how you manage them, all the things that we're talking about here. And the other thing that hits me quite a bit, we try to count how many virtual machines VMware itself is running. And it's somewhere over 25 million. It's a little hard to count at times. But again, if you just put your head around that, that's 25 million things that used to be physical servers that are now running on top of software, which is pretty amazing. And this is my favorite stat, which is we do track how often things are created. A new VM is born every six seconds or so. I'm told that's the birth rate of the US. I'm not totally sure. But the point is that this is just really growing quickly. And how we can all pull around it and take advantage of that is what I wanted to go into some detail on. So a quick snapshot of what VMware strategy is on the data center. I've been spending a lot of time on this concept lately called the Software Defined Data Center. And this is driven a lot from the idea of software defined networking. Obviously something I'll be spending a little bit of time on here. But it's incredibly exciting to think about right now in a space where we can have all of the hardware in a data center be virtualized. Obviously some of it is still hardware. But I think the notion is that you're really expecting it to be divvied out in some sort of multi-tenant fashion. And most importantly, you absolutely have to have all these pieces automated to be able to move at the right speed. So we've been really looking at this concept of how do you take all the components in a data center and make them fit this new model where you're abstracting them out from any underlying piece of hardware. You're pooling them together in some set of resources that can be doled out. And then most importantly, you are automating things so that a human doesn't have to be involved as things are provisioned and torn down and put together. So this is a huge focus for the company is really trying to work with partners and envision how you can pull all the components of the data center together more efficiently. We focus on storage and availability, networking and security obviously, and the compute side is where we traditionally spent the most time but just as much in these other areas as well. But what I think is most important for this audience is knowing the broader part of our strategy right now which is really thinking about, yes we're gonna build a software defined data center and it'll have all sorts of plugins for people to fit into. And this is really core operating system level sorts of work. But we've gotten very clear feedback as we've been growing that there are other things in the world of course than the pure VMware platform. So we spend quite a ton of time trying to make the VMware vSphere background be as solid as it can. But obviously there are a lot of other things growing up there, some of which use vSphere, some of which do not. vCloud is our own offering that builds upon this solution. Obviously OpenStack, I wanna talk about how we can be part of solutions based on OpenStack. Physical systems, I don't wanna be part of those, but they are an important part of the world as well. And then lastly obviously Amazon is out there with a growing customer base and we get just constant feedback, how can you help me deal with this heterogeneity? And that is really what I wanna talk about here today. There are three big areas that we are today investing in and actually have done some very sizable acquisition work on as well as investment inside the company. Obviously, first and foremost has been networking and security, and that is obviously the reason that we were so excited to have NYSERA join us, I'll spend more time on that in the future. But if you step back, I think networking by definition is the thing that connects a lot of different things together. So that absolutely has to be a open and heterogeneous way of talking to different components. The thing that's interesting if you look at all the things we do in the data center, most of the focus now is actually certainly making that platform as good as possible, but how can you create the tools to help people automatically use it? And that might be things that are helping you actually choose ultimately where to place things. It could be how you monitor in them. We're even spending quite a bit of time on application provisioning and where that goes. So these are some areas that at a future, hopefully open stack summit, I'll be able to share more details on how we're integrating those as well. Quite an exciting area for customers though. And then last is really something we've been focused on for a while now too, which is really thinking about this is all great, but how can a developer possibly deal with a world of this heterogeneity? We're already ending up in a world where how you write an application is too often bound to where you're running it. So I'm gonna share a few details on what we're doing there. So ultimately, this is kind of why we wanted to talk about being here today. I understand there's been some controversy over our motives for being here. There are definitely some areas that are competitive around some part of what VMware does and part of what OpenStack focuses on, but there are a huge number of areas that are in perfect alignment. And I think that's a case for most vendors and partners that are here also. So we are legitimately here to make sure we can solve customer challenges, help them deal with a very interesting and fast-changing world. And also I think that there's an interesting business for many people to be made as customers need support and different features that can fit into this model. So hopefully that kind of sets the stage. That's why we're here is to be part of this community to contribute value to it and ultimately to help our customers deal with this hybrid heterogeneous world that they're looking at. So four areas I wanna spend a little time on. These are the areas of OpenStack focus for VMware and these are the areas today that we're making investments on. There's actually about four or five other areas that we're beginning to put engineers on as well. I'll give you a little Q and A on that if you'd like, but I wanna spend the most time on these four so you understand specifically what we're doing. So let's start with Nasirah. I think most people here are fairly familiar with Nasirah. They've been obviously doing a lot of interesting things from the start, very involved with quantum and really getting in the middle of how we're thinking about networking in this new world. For us, this was just an outstanding team of engineers thinking about the next way that networking will work. I always remember my first meetings with the team where they talked about networking as a distributed systems problem as opposed to a pure networking problem. And I think that sets the stage for what a lot of us care about, which is how do you manage these things to scale. The other thing that I think was very exciting for us was this starts with how do you make networks that can be very flexible, very virtualized, but in the same way that we got started in servers and storage on virtualization, it opens the door for really thinking about all the other challenges. And one thing you can see in our booth here if you're interested and also something you'll hear more and more about is we think we can collaborate with this community to open up APIs and really set the stage for higher level networking services to fit into this model quite well. And if you go around and you talk to really the big enterprise customers, I spend a lot of time doing this, asking them where the time goes. We made it very easy to spin up a virtual machine fairly quickly, but when you're in an enterprise it takes another five days to get that virtual machine properly plugged in on the right VLAN with the right firewall with monitoring tools plugged in, updating your ERP system so it knows where it is. This next five days are the things that we all need to address to make people more efficient in an enterprise concept. And so a lot of what we're gonna be focusing on is how can you get things like IDS systems and firewalls and load balancers really to just fit squarely in the same model. So anyway, if you wanna learn a little bit more about that, we'll have sessions when the booth is going into it as well. But the other thing, you know, the real reason that we acquired Nacera for a very lofty price was because of the core work that they've been doing with this community itself and making sure that they could fit into this heterogeneous world. So very excited that Dan Winlet is probably here somewhere, right? Dan back here, obviously a lot of you know him as the PTL side of things, obviously really helping us understand the community and how we can best work with it. We're 100% committed to continuing to work on Quantum as well as the open V-switch. And we would just be silly not to do that. In fact, we're gonna be doubling down on that investment. So just be very clear, we wanna be part of this community on the networking side and do so in a community fashion. The other thing that's pretty exciting is as you listen to a lot of the clouds that are out there already, you know, by nature of Nacera being part of VMware, this is now a set of customers of ours that we care deeply about making successful and working with them. So a lot of you have seen different sessions but whether it's eBay or NTT or Rackspace with a great offering or AT&T recently, as well as several others that haven't yet been announced, you know, these are customers for Nacera and these are things that we are committed to making work very well. So we're very excited about that, learning about what's going on there. And part of that is learning more about how a large open stack development process works as well as a deployment process. What's also exciting is that not many people know this but VMware is now running ourselves a very large open stack offering by nature of Nacera who's been running a production open stack lab within their own group about 1500 VMs I think is where we are in that front. So a fairly sizable effort that's running now and we're getting all our learnings from that as well. So the next thing I wanted to talk about is Cloud Foundry and maybe just a quick show of hands. How many of you have heard about this before? Oh, super, okay, well, that's good news. It's something that we've been working on for about two years now. It had a birthday recently where we really called out what was going on here. The idea here is actually pretty straightforward. You know, if you really want to leap forward, if you're looking at an enterprise today, they write applications pretty much in Java or .NET and they're writing them in a way that has to deal with a lot of their legacy setups and they care deeply what version of the database is in place, they have to know what version of the Java container is in there, they prescribe all these things and they have to deal with versions as they change. I think anyone who's been in development knows that that's a pain and certainly from a cloud perspective that's something that really slows you down. So this is really around the notion of platform as a service as most people have been referring to it. Cloud Foundry is designed to be a polyglot, I think is the word you use now for a multi-language system. So certainly you could have everything from PHP to Java to different languages. We even have .NET on top of this now so that people can choose which language they want to write in in the same common platform. It's also multi-service in the sense that we want to have it very easy to plum into particular database services, different blob services or different add-on services that a lot of public cloud vendors are looking to add, value add that they can have developers take advantage of. So figuring out a way to make this very easy to consume these add-on services and a way to do it in a way that doesn't fully, fully lock you into just running one place is a real challenge, but that is a big focus here. And last but certainly not least is that if we really want to make this work, what a developer needs to know is that they can write an application and as I said at the very start, what you write an application in or how you write an application should not dictate where you run an application. And as you look at some of the PaaS offerings that are out there today, the only place you're ever gonna run it is in the particular cloud where that lives. And we don't think that's the right model. So anyway, we launched a very big effort around Cloud Foundry and are investing quite a few people in this. 100% open, it's on GitHub, you can go download the latest right now, Apache license from the start. And in fact, on the very first day that we showed Cloud Foundry, it was actually running on Amazon. So it is absolutely the case that we wanted to run anywhere that people can run it. We've done a very nice shimming layer so that it's very easy to plug it in different places. We even let it run on your local laptop. It can run physically even. So this is really a pretty interesting area and I'd invite you to take a look at it. The other thing I wanna call out though is that this works great on top of OpenStack. And specifically I wanted to thank our partners at Piston who spent some time early on looking at how to make this fit directly in their distribution. They've open sourced the software that's there and I believe you can go learn a little bit more just outside at the booth that's out here and learn about what they're doing on that front. But anyway, the real focus here just to be super clear, we think the world will benefit quite a bit by having a way to write applications that's easy and productive and that does not tie you to any particular place to running it. And I think that fits the mission of this community extremely well. And I'd invite you to go to cloudfoundry.org where you can download the complete distribution, work with it. You can also go to cloudfoundry.com which today is a free service that lets you kick the tires and if you've never written a Sinatra Ruby app, here's your chance. You can just log in real quickly, try it out. So that's the focus for Open Platform as a service and something we think could be a very interesting way to allow the OpenStack community to have a higher level offering around how you write applications. Third area is a little bit in the plumbing and something that a lot of you are familiar with. So we also acquired Rabbit MQ which is really focused on a message bus that a lot of people use. I think those of you that have worked deep inside of OpenStack know it's part of a lot of these deployments. We have a big effort within our team which is based in London mostly really trying to make sure that we can satisfy both the scale, HA, and performance needs of its use case that's being done here and without going into the specifics I think if you talk to a lot of public cloud vendors that are here, they're looking at this core messaging bus as a way to do these add-on services and really do some value add around their OpenStack developments. So again, the point here, this is a commercial open source. This is a tool that is very interesting for developers in their own right and we're committed to make it be a great infrastructure for those of you building scale systems on top of OpenStack along the way. Last but certainly not least the question that I get asked probably the most so far which is what about vSphere? So as you can tell, obviously this is something very pervasive in the enterprise and we think it is obviously the best and certainly the most popular way that enterprises are running virtualization. So of course we've been asked quite a bit how can we have vSphere be part of the overall OpenStack offerings if I wanna use that or if I already have that how can I make the rest fit around it? And I just wanna be real clear we've been talking about this quite a bit but this is the first time when we're able to come out publicly and say what our goals are here. It's pretty simple. If a customer is using OpenStack we would like them to be using vSphere. We think it's the best way to do it and we're gonna be investing quite a bit to make sure that it runs well. And so that's the other interesting news today which is we're enhancing the VMware Nova Compute Driver to support vSphere and its capabilities. And I think that's something that's extremely important to call out. There is a couple of different things. There is a core hypervisor that just runs VMs very simply. It's rather dumb to be honest and all it does is create VMs and run them. But there also is a whole lot of value to get out of virtualization platforms around HAs, around SLAs, around live migration for the right reason, around power savings, huge areas of investment. And so what we also wanted to do is make sure that when running in an OpenStack environment you can take advantage of these higher level capabilities if you'd like to. So this is a very big investment for us and something that's quite important. I think a lot of you hopefully saw this advertised. We'll have Sean Chin here who's talking about specifically what we're doing and he can give a lot of answers around how we're plugging it in and what we hope to accomplish there. But this is obviously a very interesting thing for us. Again, we want people choosing vSphere and using it and if they choose to run it in an OpenStack environment, great. If they choose to run it in one of the other environments including custom cloud management systems we want to enable that as well. But that's where we're coming up from this angle. So that's kind of just the beginning. I wanted to spend a fair amount of time in a moment on the question and answer session but I just kind of wanted to reiterate what we're trying to do here. So we are getting a lot of feedback from customers and we have for a long time that we're moving to a world that is really a lot of hybrid combinations going on here. Hybrid means a few things. It could be that they're running some things on premise and that they're running some things in a public cloud. It could be that they have a heterogeneous pools of infrastructure where they have a big VMware cluster. They might be trying Hyper-V for something over here. They might need some things to be running physical and we're being asked quite a bit to not only recognize that that exists but actually to add value and make it simpler. And so that is specifically why we've been making some very large investments and acquisitions. And then lastly, we're really trying to make sure that people have the choice into how they want to run things and we find just constantly, if we're able to give them choice and show value in different areas, then the market will decide in many ways. So I believe there was supposed to be a sheet of paper but there are a number of other sessions that you can see as well as going to our booth. Actually, I have a little extra incentive for you to run by the booth if you'd like. A lot of you probably use a Mac to do your development and so I'm able to give out several hundred free copies of Fusion because we want developers using Fusion which is our tool that lets you run Linux on top of your Mac and really develop for OpenStack. So I think we have 300 copies of free Fusion out at the booth right after this. Stop by and grab it and hopefully do some great OpenStack development there. And then I also want to call out, we have an email alias that you can use if you have any sort of comments whatsoever on things you'd like us to be doing, things you don't think we should be doing, any feedback whatsoever, you can send email to openstackatvmware.com and we'll get back to you on that, on the feedback from whatever questions you have. So hopefully that gives you a pretty good flavor for what we're doing. I thought I'd open up the floor for five to 10 minutes of questions right now and we'll go from there. And I don't know, do we have a microphone that we wanted to hand around? Oh, it's sitting right in front of me. Morning, you mentioned support for Open V-Switch. I was wondering if you're going to be supporting, replacing the standard VMware Switch with Open V-Switch or will the standard VMware Switch become Open V-Switch? Yeah, good question. So those of you that don't know, in the core vSphere product that we have is a distributed V-Switch. We've actually, if you're not aware of it, we've actually made API to allow that to be pluggable by different areas. Cisco, for instance, sells something called the Nexus 1000V, which plugs into that same area. They particularly do it so that people who want to write to the Cisco CLIs or have all the feature set that's in there can be leveraged. So we are looking quite seriously at what aspect of Open V-Switch to merge and to have completely interoperating within the vSphere environment. Kind of we haven't decided yet because we're just kind of getting through the acquisition. No matter what, we want to make sure that there's very clear interoperability from an API and from a feature set standpoint. So that's kind of the low end. How far we go from there is something we haven't decided yet that we're definitely open to looking at very broad incorporation of it. Very good question though. It's something we talked about from day one when we did the acquisition and the idea of having some common way of managing across the pools that might be well accomplished by doing that. What's some other questions? It's awfully quiet in here. Yes. Now the question was, what is VMware doing on the storage side of things? I'd actually love to talk for hours on the software defined data center concept but maybe I'll give you a few quick tastes of it. So as people get started with virtualization, a lot of what they're doing is existing enterprise applications. A lot of you thinking about the new world think often about blob stores and of stateless applications which are super interesting and super valuable. A lot of our focus has been for existing stateful applications that are doing high bandwidth communications, being able to interact with storage well. So a couple of just kind of examples there. The core thing that we do really is in a multi-tenant environment making sure that people can consolidate virtual machines but give storage level guarantees. I think I heard Barry from HP talking about production applications. Reason number one, two and three to get applications going there is to get guarantees around whether it's security or performance. So a big focus is in a consolidated environment, how can you guarantee storage performance? Now the second thing that is really core to a lot of what people look at now is when you really spend time in a broad storage environment, there's really two worlds. There's the world of distributed DAZ storage where each server has disks and then there's also the network attached storage. And we spend quite a bit of time trying to let people move around between the two of them. So a lot of our focus has been on shared storage and allowing people to do failover and multipathing. A lot of it also has again been focused on how can you seamlessly move around the data center so we can migrate between arrays without taking them down. Huge part of the focus for people that are running existing enterprise applications. Going forward, we're spending quite a bit of time looking at embracing more of the storage model where it is a bunch of individual disk drives attached to storage. So we've announced an effort to do, we call it VSAN, which is really building a SAN out of a collection of disks that are spread out across servers and really making that highly available. But for me, what I really love about it, it's a concept that people here would be familiar with. The idea is you plug in a server and the compute resources and the disk resources simply join the pool and can be divvied out automatically. That's very key for one other thing I wanna spend just one minute on, which is really looking at next generation applications on the same platform as existing applications. So I put up a number earlier on that said 60%, maybe it's 50 to 60% of all server apps are running virtualized. The immediate question should be what about the other 40, like what's not going there? So a big chunk of that is around enterprise applications as people wanna be more familiar either organizationally or technically with how they run. And then a big portion of it is new applications that today are creating clusters, not running on a cloud, not running on any of the things that we're creating. You see this on Wall Street where they're doing low latency applications. You might see this in areas such as big data. I can tell you how many people I go to now and they're building just for test and dev, they're gonna build a small cluster to run big data workloads. Those are the things that in the future become very large clusters and they're gonna be treated separately from everything else. So a lot of what we're trying to do is make sure that this common platform, in our case vSphere, can be the place where workloads of new types and old types can run equally well and add a lot of value. And so a core part of that, I'm getting around to the final question, is how can you leverage local disk latencies, especially for things like big data applications, make sure that you're getting the full performance while still getting all of the abstraction and other capabilities as part of it. So a big part of our focus is on that whole front. And ultimately I think to have the software defined data center, it means you have no knowledge or requirement that you know that you're on a particular array. And so we spend a lot of time on that. Yes. I think the question, would you allow OpenStack to manage vSphere as opposed to the other way around? Is that how you ask the question? Yeah, so that's certainly the idea of this compute driver is that if you wanna use OpenStack on the top end and then have this be the kind of the compute aspect of things, the whole concept is that we will plug into that model with vSphere. I'll reiterate the point again, we think it's very valuable to be able to not just use it for core virtual machines, which will be fine, but also we wanna make sure that managing it in an OpenStack environment, you can take advantage of some of the advanced features as well, so you can go to the lightning talk and hear about how that works along the way. Yes. I'm sorry? The plan for OpenFlow? I'll probably defer you to talk to the NYSERA team on where we'll be heading with that. Is there a session on that later? Or something explicitly you wanna go into? So sorry, Dan just said, feel free to contact them afterwards. I'm not sure we have any specific session on where that's headed, but that'd be a good way to reach out and really get in the middle of it all. Yes. So we're pretty early on in the overall planning. So there is a lot of overlap between things that the broader vSphere picture do and things that the OpenStack world is headed towards. You know, our phase one is walk, so let's make sure that, however you wanna launch jobs from the OpenStack world that they can land on here and ultimately use them, but I think you could do it at the next level and I'll use this for just a quick moment to call it another area that if you invite me back to the next summit, I'll hopefully be talking about, which is we bought a company called Dynamic Ops. Maybe has anyone here heard of Dynamic Ops before? Okay, super. So the concept here is around very top level placement of workloads. So there's all different layers you wanna place things, but Dynamic Ops is used very heavily starting in the financial industry. It was a spin out from Credit Suisse, and it's really used for a top level company when they're making decisions about workloads. And I think this is really important to think about from this community standpoint, as well as ultimately where you're having things making decisions. At a very top level, you have a company, they have a bunch of workloads that they wanna run, maybe it's in some sort of app catalog, and they have constraints. So it might be this application is a test and dev workload, and you should never spend more than X dollars running it, where it might be this has sensitive data, it can only stay in New Jersey, don't ever move it somewhere else. And you can imagine a lot of other things that are on there. And something at a very top level needs to choose from the time someone comes into the portal and says I wanna run an application, ultimately where it goes. And so there's a lot of workflow that might be involved with that, as in this type of workload needs approval from these different people, let's send emails. It could be inventory updates, so I need to update my ERP system to say, I'm now running a new thing, update license count, whatever it might be. And then ultimately what we wanted to do is choose from a lot of different pools of infrastructure, and then hand off control to them. So it is a top level decision maker, it might say this is fine to run in this open stack based public cloud, this one's extra sensitive or for whatever reason it needs to run locally in our private cloud. Anyone talks to a physical hardware, so it might even say, this one you don't wanna virtualize, let's send it here. So we're investing at that layer quite a bit, and again very heterogeneous, so it could choose any public or private cloud based solution. And then we have quite a bit of effort once you're in the vSphere environment, in terms of to get your SLAs, where do you wanna place things, to make sure that you have anti-affinity rules, so if one piece of hardware goes down, make sure you know where to place things. So we kind of have this sandwich of placement, and I think there's a middle level placement that we can integrate quite nicely with as well. Long-winded answer, but I hope that gives you a little flavor for what we're looking at. I think we have time for one or two more questions, yes? The question was, can you run Cloud Foundry within your own data center? Absolutely you can. We don't have any commercial offerings at all around Cloud Foundry, although a number of partners have spun up to do them. Technically, Cloud Foundry is a very cool system if you're into like scaled architectures. So it is a set of virtual machines that auto deploy and auto scale out and take advantage of whatever you have. It's not super tightly integrated with any particular products that we have, but a lot of people are kicking the tires on it directly on that. It could run on, I mean, certainly on OpenStack or anywhere else as well, but there's no commercial offering around a private Cloud version of that today. One other question? Yes? Boy, that's a great and hard question. It was, how do you see, did you mean VMware or VMs? VMs were specifically in that kind of event. Okay, yeah, this is one of my favorite topics, so I'll try to be somewhat terse in the interest of time for everybody. So the fun thing right now is that as we've come in across the industry and changed the way that things are being placed and that they're moving all the time, we've kind of screwed up a lot of models. My first five years at VMware were apologizing that your CMDB no longer worked because the guaranteed correlation between a workload and a server was broken. So a lot of things have been broken by the nature of consolidation as well as movement of applications along the way. And I think security has both that challenge that things have changed and an opportunity as well. So the challenge is things that you've been doing in an enterprise, again, I'm not sure how many of you have really set this up, but if you go into an enterprise, there are appliances of all sorts and they're set up in sort of tree cascaded model where you put your application behind the firewall or behind the IDS and you are restricted where it can go because of this tree of security. It leads to a couple of challenges. One is that device typically has to have very high bandwidth. It becomes the choke point in many ways for things coming in. It also, almost by definition, sometimes has to have a lowest common denominator security because everything behind it needs to kind of satisfy some of those rules and there's exceptions, but what we've been seeing over the last three or four years is if you leverage virtualization, you can actually do two things that are really nice. One, you can actually break up these edge devices in many ways that used to be very large appliances and you can break them up and sprinkle them out through the network, which allows you not to have to have peak bandwidth needs for them, but what I really like is it allows you to bring security closer to the application or the workload that it's actually doing. So we've invested quite a bit around network interfaces that allow these different devices to literally come and wrap around a virtual machine. So you can do tighter security, but more importantly, it actually will then follow the virtual machine wherever you go around with it. So there's a class of things going on with that. That's actually one of the core areas we think with NYSERA and VMware and OpenStack, we can do some pretty interesting things around interfaces to allow these different devices to fit in. There's actually other cool stuff going on within the virtual machine around automatic check pointing, around offloading, virus detection, a lot of other security areas. And then the third area of security of importance for a lot of people in this room and all over the place is more on the compliance side of things, which is less fun security to talk about. But it is, how can I guarantee that I will pass an audit that says that I've had the right control over infrastructure, that sort of things. So we spend quite a bit of time in this management and automation area also allowing things that can check and confirm that. And those are assets, as I was saying, I think we can bring to this community also over time for the customers that want it. Let's do one more question. Yes. Yeah, the question was, I think it was a little bit turned around. It wasn't vSphere managing OpenStack, it was OpenStack managing vSphere running on top of vSphere. And the question was, how does that correlate to vCloud? I think is the way to, is that how you're asking it? So there are definitely areas of competition between a product we have called vCloud Director and some aspects of what OpenStack does. We have a few things that are going on. One is they really satisfy different goals on a bunch of fronts. OpenStack has a very different extensibility model and is obviously very popular for a bunch of fronts. So we wanna embrace that side of things. vCloud Director is very tightly integrated with our platform and enables deep creation of virtual data centers and a lot of the software-defined data center concepts that we're doing. So the bottom line is that there's definitely some level of competition between the two. There's also a level of interoperability needed between the two. So we're spending quite a bit of time on the APIs that talk to what we call vCloud and seeing how we can make those very interoperable between how people are talking to OpenStack at the same time. So we think the two need to interoperate, but the core message should be that we wanna provide choice. So if customers are gonna use, they'd like to use the things that vCloud has, absolutely will help them. If they like to use the things that OpenStack has, absolutely will help them as well. So I don't know if that totally answers it, but that's the state of the world there. Okay, well, thanks very much. Again, we have free copies of Fusion out in the booth. Hopefully some of you would like that. And if you have questions, also go to the booth as well as the lightning session we have in a little while. So thank you.