 Okay, so I'll go ahead and get started, as people are still rolling in. My name is Boris Ransky, I'm a co-founder of Mirantis, and today we'll be talking about making money on OpenStack. Judging by fairly thin attendance, I can say that this is really a problem that bothers the community very deeply. But I'd like to start with a little bit of a detour and talk about my credibility with respect to ability to speak on this matter. So as many of you might have guessed, I originally am from Russia. I was born in communist Russia, back in 1981. And if you were to come to the communist Russia back in 81, this is kind of a typical picture that you would see. So what it is, is a piece of crap Russian car called Lada, a marvel of the Russian automotive industry. It barely drives, probably eight times out of ten it wouldn't even start. And the reason I decided to put it up there is to just kind of, you know, I think it somewhat symbolizes the quality of life that people had to kind of, you know, sustain back in those days. And you know, naturally nobody had any money, lines to buy bread, things like that. So naturally when in the early 90s, a lot of opportunities opened up after communism fell. They jumped at the opportunity and the way people did it is they weren't really so much bothered about, you know, building a disruptive product or, you know, changing the world. People jumped at the opportunity of just making money quickly. The shortest path to most money is what most people pursued. And now if you come and visit Moscow today, what you will see is this. And I mean, to some extent I think, and this is actually a real picture and I don't think that's anything, it's not even special. If any of you have been to Moscow, you will actually see a lot of these things. You just need to roam around the city for a couple of days. You know, to some extent it maybe is a testament that sometimes this kind of mentality can work and can be effective. And I as somebody who's been, you know, exposed to this transition and as a Russian person and generally have naturally inherited some of that, you know, I want to make money fast type of, you know, approach to dealing with things. And on a completely unrelated note, I'm a board member of the OpenStack Foundation and as such it's my duty to, you know, attempt and evangelize OpenStack around the world. And when you hear most people trying to evangelize OpenStack today, you know, the common words that I hear coming out of their mouth is, you know, it's a great community. Let's get together, let's build the best open cloud ever. Let's, you know, get together, do kind of meetups and drink beer and write Python code. And when I listen to all of this, I kind of, you know, I imagine this immediately, which is just a bunch of nerds writing Python code, which for me as a guy that comes from Russia, with that money mentality is not very motivating to contribute anything to OpenStack. So what, you know, the kind of a point of this whole presentation is to motivate everybody to look at OpenStack for a slightly different prism, and that is a prism of making money. So when I think of OpenStack, I think of this, that stacks of cash that I can make leveraging OpenStack ecosystem. So now, hopefully I've convinced you that, you know, I'm a credible person to talk about it. Let's dive into it. So there are four key questions that we need to answer that effectively, you know, are relevant to making money on OpenStack. The first question is, is there money in OpenStack in general? Second is, how much money is there? Why OpenStack potentially is a better vehicle to make money than, you know, the alternative stacks or alternative technologies that you might want to kind of invest in, and naturally the most important question is how is it that you make that money? It's really, I'm not going to spend a lot of time on it. It's a simple question, you know, if I was presenting at the beer convention, probably I would talk about it a lot, but here most of you probably already believe that this is the case. Anyway, we can run through the logic really quickly. The logic goes, something's false. Cloud is big, a lot of money getting invested in cloud, everybody knows it. Cloud is open, a little bit trickier one, but we'll talk about it. And third, OpenStack is the king of, you know, the open cloud ecosystem, which if you put all of these premises together, ultimately translates to, you know, probably a good opportunity being there. So cloud is big, everybody knows it. This is actually the data that I pulled off the cloud scaling blog. RainDebias did the analysis of the Amazon Web Services growth and Rackspace growth, 100% year-on-year growth in infrastructure as a service clouds. I mean, these are kind of impressive statistics, a lot of money being poured in, everybody writing about it, no doubt it's big. Cloud is open, so this one is a little bit more interesting because it's a little bit less understood. So when it comes to kind of a cloud paradigm, I think that we're in an interesting territory because if you look at the application infrastructure, kind of an industry's history, the general trend historically has been that open and open source has been just, you know, a cheap alternative to proprietary. Typically the way the industry would move forward, somebody would come up with some, you know, disruptive technology, they'll disrupt the market, they'll make a lot of money disrupting it, they'll corner the market and then some open source alternative will appear that is probably less feature rich and it's really kind of an attempt to, you know, recapture the market that's already been created by somebody else. When it comes to cloud, the picture completely flips because cloud is kind of a baby of the open, to a large extent. I mean, cloud has essentially been born, the term cloud has been pioneered, I think, as we use it today in reference to infrastructure clouds by Amazon, but the idea of cloud as a lot of people in infrastructure space think of it today is really kind of a byproduct of the, you know, consumer internet revolution. A lot of companies that had to deal with running applications at massive scale had to completely rethink how they build out their infrastructure stack. They understood that, you know, if you were to go, the traditional enterprise approach, the license fees that they would have to pay to VMware and Oracle would, you know, far outpace the revenue or money that they could ever make. So they built this completely new kind of open infrastructure. It's either proprietary things that they open sourced or the open sourced things, existing things that they pulled in. Ultimately, the bottom line here is that they don't pay license fees to anybody. And they've kind of paved the path for the cloud and everybody followed. Amazon, you know, actually opened up their, you know, infrastructure and started selling it as a service, which is where cloud came from. And this is kind of the cloud. And the third premise, again, a lot of people don't like when I talk about it, but I am of a subjective opinion that OpenStack has won the open cloud wars. This is a slide that Josh McKenzie from Piston likes to show, average number of developers in an IRC channel. This is data of June 18th of this year. But the point is that if you look at just the sheer size of the ecosystem, the number of people that are attending the OpenStack summit, the number of people involved, it is substantially larger that any other ecosystem that is kind of in the same space. Another statistic is the just size of the, you know, the foundation, the amount of money that's been committed by the sponsors to the foundation. It is the second largest open source foundation only after Linux itself, which operates at, you know, OpenStack Foundation, 6 million a year operating budget Linux is 9.6. So let's try to tackle a different question, a more interesting question, that is how much money is there in OpenStack? So the first thing to look at is actually, you know, what do we mean by money? Money is spent on infrastructure in different ways. Service is hardware and software is where most of it. This is kind of a pie chart representing the traditional IT enterprise, enterprise IT spent as it relates to those three categories. So if you look at any kind of sub-segment of the enterprise IT market, there are kind of components that are split between services, hardware and software in there. So there's many ways to analyze it, and I'm not going to try to convince you that my way is the best, and I think that the point of the exercise is not even to get to the exact number, the point of the exercise is to kind of ballpark it, to understand what we're talking about, you know, million dollar, 10 million dollars, billion dollar, trillion dollars. One way to look at it is to look at the existing market that's there. So kind of OpenStack is in a new path, but we can try to ballpark it by looking at the market that OpenStack is displacing. So that market right now is largely dominated by VMware, at least when it comes to virtualization, there we go. VMware owns roughly 60 percent of the market. If you look at VMware annual report, they have a line item there that specifically says cloud infrastructure licenses. So that's how much licensed software they sell that goes into the cloud infrastructure bucket. That's 1.6 billion for 2011. If that's 60 percent of the market, we could say that maybe 2.7 billion is the total license component of the cloud infrastructure market. And if you were to add services and hardware to that, leveraging the pie chart from the previous slide, you can ballpark it at about 15 billion. Again, the point is that, you know, nobody really knows how big this market is, exactly. The best we can do is the ballpark, but everybody kind of understands that the market is very nice. So why is it better to spend your efforts trying to make money on OpenStack than on Eucalyptus or CloudStack or VMware? Those are fairly viable ecosystems as well. VMware is definitely a much bigger ecosystem, so why OpenStack? The answer boils down to the fact that, unlike all the other ecosystems, OpenStack was kind of, you know, put together by its original founding members with the idea to create the ecosystem. Instead of creating just a piece of software that is open source, that is going to be backed by someone vendor, it was envisioned as an ecosystem from day one. And the idea is that it's going to be an ecosystem that will have kind of an even balance between a lot of participants. There's not going to be any one vendor protruding and nobody is really going to be the second citizen to somebody else. It's started by the community, it's being marketed jointly by the community and all the participants. And really the goal of the ecosystem is to enable all the participants to kind of make money on that ecosystem. If you look at the alternatives, I mean, they're all about the single vendor. So they're all started by a single vendor. The goal is to create the ecosystem around which the single vendor can reap the benefits and it's impossible to be, you know, a Eucalyptus player yet be better at Eucalyptus than Eucalyptus. And it goes the same way for CloudStack and it goes the same way for VMware, so it's much harder. And the idea behind their, you know, the idea behind even open sourcing some of these products to begin with, in my opinion, is not so much about creating the community but about kind of, you know, throwing this veil of no vendor lock-ins. A vendor lock-in is this issue that a lot of people are concerned about. You know, I want to have no vendor lock-in and a lot of people just kind of tend to associate no vendor lock-in with open source. So if you're open source, no vendor lock-in, you're good. So, you know, if you take Eucalyptus or you take CloudStack, you make it open source, you make it even Apache, even better. It's open source, definitely no vendor lock-in and that's the end goal, less so to really create the ecosystem. And that's just my personal opinion. So now the most important and interesting question and that's how do you actually get the money that's out there. So you can kind of split this up into, you know, a series of sub-questions and the first question is who do you actually sell to? To be able to make money, naturally, you have to understand who your customer is. And for OpenStack, this kind of changes as the ecosystem evolves. So what I tried to do is I tried to put together the simple graph. On the y-axis, we have kind of the risk of adopting OpenStack and by adopting here, I'm speaking kind of of adoption in a broad sense that can either be, you know, deploying it if you are an enterprise or it can be actually, you know, starting to try to integrate your products with it if you're an infrastructure vendor, can be a lot of things but basically touching OpenStack. So what is the risk? And also surprisingly, the risk over time goes down. I specifically plotted the time, 2010, when OpenStack was first started and 2017 is kind of a long enough horizon for us to kind of be able to see the whole cycle of adoption. So the first year when OpenStack was started initially, it was a lot of, it is true, it was a lot of vendor-driven hype. It was a lot of vendors that felt that maybe they're a little bit late in the cloud game that, you know, attaching themselves to the OpenStack bandwagon is a good strategy and they started investing a lot of money in integrating their existing solutions with OpenStack in some way so that as OpenStack ecosystem develops and there is market pool, their solutions get pulled in. Some companies just, you know, would build drivers to, you know, to have storage work with novel volume. Some companies would build an entire strategy that would be kind of more all-encompassing but the bottom line is it's infrastructure vendors like HP, like Cisco, like IBM that started spending money on doing something of OpenStack and that's really been kind of the exclusive trend for the first year, at least as we at Mirantis have observed it. And a lot of our early customer base is exactly from that segment. Now, the second part of it is service providers and internet application vendors and this is where we are today, I think and this is where we will remain for a little while. So we'll go over kind of the motivations of each of these, you know, potential customer segments in subsequent slides. And the third, and I don't think anybody argues of it, the most conservative adopters is the enterprise. I personally don't think that, you know, there's been, you know, any significant adoption of OpenStack or any cloud platform, so to speak, in enterprise at all whatsoever right now and I think that we're looking at the horizon past 2014 for some real adoption to take place. So, you know, if you try to layer kind of the sizes to this subsequent, to this kind of market segments then naturally the infrastructure vendors are going to be doing most of the stuff themselves. Maybe they'll buy some services. They spend a little bit of money tackling a little bit more of a strategic problem for the service providers and internet application vendors. So a little bit more money and the common thinking is that it's the enterprise that's when the big bucks start crawling in. So let's look at each one of these in a little bit more detail. Infrastructure vendors, what's the use case why the infrastructure vendors are really spending money on this? Like I said, to basically not get commoditized and to ride the cloud hype. You want to differentiate around cloud. You want to integrate with OpenStack and that's really the use case that a lot of them are kind of trying to solve for. So what is it that you can sell to each one of the segments? So if it comes to infrastructure vendors, the only thing you can really sell to infrastructure vendor is services. It's unlikely that you can kind of bundle your OpenStack and sell it to infrastructure vendor. It's unlikely that you can sell OpenStack support to somebody like HP, for instance. And the original kind of the first year wave is pretty much exclusively services driven. As we move into the service providers, internet application vendors, there are two different use cases. For service providers, another one for internet application vendors. But I bundled them together into the same bucket because I think that from the standpoint of adoption for both of these categories, it is happening at the same time. So for service providers, naturally the use case is, hey, cloud is big, Amazon found a part of gold. We want to do the same, let's build Amazon, let's use OpenStack for it. So that's kind of the general bottom lining for service providers. For application vendors, most of the use cases we've seen so far are kind of twofold. One is DevTest scenario. Build the cloud to a kind of continuous deployment environment which would integrate with OpenStack and you can do your development more efficiently. But the more interesting scenario and this kind of maps very well onto the Webex case study, which is Webex is the second largest SaaS vendor in the world they were presenting. And that bucket is trying to replace kind of their spaghetti-like infrastructure that has hazzardly evolved over the course of the company's evolution with something that is more or less standard. So not every SaaS vendor or internet application vendor was as smart about building their infrastructure as Google, for instance. So they're kind of gradually evolving it, patching and stuff like that as their solution was getting adopted and a lot of them today have ended up with this kind of very weird kind of spaghetti-like infrastructure that is not very central. They have pieces of VMware, pieces of other open infrastructure solutions and they have to kind of develop and build their application-level logic against this weird infrastructure. It's very hard to maintain and OpenStack is potentially a standard that can replace it. So let's kind of do a project by step replace all of this weird infrastructure of something that's going to be one beautiful and elegant solution. So what can you sell to these guys? Again, I think that most of the stuff that you can sell to these guys is services, but we are seeing the emergence of some possibilities for selling some of the products. You would not be able to package OpenStack or sell your own distro of OpenStack, in my opinion, to a service provider or guys like WebEx. But you can kind of find some components that might be of value to those organizations, which they'd be willing to pay subscription for or kind of a support-like model that will solve things for them. So we're kind of starting to see some emergence of opportunities to build a leveraged business model at this stage. And finally, the enterprises. Again, I don't think we're really there yet. I think we're just scratching the surface at this point. From the standpoint of the enterprise, the use case for something like OpenStack naturally is embrace the new paradigm of open efficient infrastructure versus the old paradigm of database automation infrastructure that is largely powered by VMware. So displace VMware is really the bottom line. And what you can sell to enterprises, is pretty much everything, because unlike for the two categories that we talked about in the past, for enterprises, for most enterprises anyway, the IT or application infrastructure component of IT is rarely seen as a core competence and they're willing to pay money to third-party vendors to solve their problems for them in that space. So you can sell engineering services, you can sell managed services, for instance, like what Rack Space is trying to do with their private cloud model, just build a cloud and will operate the cloud for you inside your firewall. You can sell support subscription like Redhead does it, you can sell a whole bunch of tools around such as monitoring tools, deployment tools, et cetera, et cetera. And you can sell appliances and hardware, so you can imagine selling something like a V-Block only running OpenStack, for instance. I'm sure that enterprises might be interested in that. So the question that I have, and I don't have an answer for it, but the question that I have and I think that's worth pondering is that in the earlier slide we made this assumption that the money is in the enterprise market. But we see two trends. First trend is enterprises being very slow at adopting the new paradigm of infrastructure, the cloud paradigm, and we consciously pushed them off the kind of 2014-2017 time frame. And the second trend is that as time passes what we're observing is that enterprises are really less and less interested in really spending money on building out their own internal IT infrastructure. As it pertains to the infrastructure layer, definitely. So the good trends to look at is just the complete obliteration of the on-premise enterprise software with the SAS model. So on-premise enterprise software is dead. Everybody in the enterprise software market agrees with it. It's simply not growing, and SAS software is just taking off and expected to grow even faster. And given that about 50% of the enterprise infrastructure spend is somehow attached to the enterprise software spend, ultimately that means that enterprise is spending less money on building out their own infrastructure. Another trend that is kind of puzzling or not puzzling but pointing in the same direction as the adoption of AWS infrastructure as a service solution. And if you put these two trends together enterprises being really slow and being at the tail end of adopting cloud infrastructure. And the other trend is this tendency of enterprises to spend less on their internal infrastructure to begin with. The question is really how much money is going to be there in that market when enterprises are finally ready to adopt cloud-like infrastructure at mass. And again, I don't have a cool answer for it but I think it's something worth thinking about. So moving on to the next section let's talk a little bit about the ecosystem. We've covered the theory part of how potentially one could make money in open stack. What we're going to do is to look at how other companies are trying to make money in open stack. So I like to think of it in terms of scale. If you look at the entire open stack ecosystem you can align almost all companies along a certain scale. And on one side of that scale you will have companies that are trying to kind of attach themselves to open stack ecosystem but on the other hand their message is more of a traditional kind of a proprietary message. So they're easy to adopt. They're kind of out of the box, just take me push the button and it magically works. But there's a lot of proprietary components and you maybe have to buy hardware licenses. On the other side of the scale are the companies that are kind of more embracing the open stack momentum, the open part of it and saying, okay, open stack is a platform maybe we'll sell you services maybe we'll sell you some tools but we're not going to be selling anything proprietary. So if you look at the existing ecosystem you can have these four camps. There's hardware camp, software camp, support camp and system integration camp. Hardware camp largely kind of airing on the proprietary side system integration services camp being on the open side of it. And the players align something like this and let's just move forward and examine each one of these subcamps in a little bit more detail. Hardware camp. And I'm not making a statement that one camp necessarily is better than the other here. I'm just saying that they have a different properties and all of them will map differently in terms of their ability to sell against that open stack maturity curve that we looked at. Hardware camp, it's very much aligned with enterprise buying patterns. Enterprises like buying solutions that are out of the box and complete and a kind of simple and push button and there's a big existing market for it. You can sell for a channel, you can build a leverage business model so the upside of being in that camp is fairly sizable. The cons of that camp is that it's very hard to penetrate the market because enterprises are used to buying from their incumbent hardware vendors as it comes to infrastructure. It's known to be very segmented and it requires a fairly significant R&D investment upfront to really put together a piece of hardware or software that runs and just a lot of things they have to put together to make them work and make them work nice. Some of the key players in the camp Nebula which maybe most of you, I'm sure most of you know their positioning is, you know, we are the iPhone of the cloud so they're like the Uber guys on the Uber proprietary side. They're, you know, it's like a magic box that runs some magic software that is somewhat open stack and you plug in your infrastructure to it and your orgasm as you see it turn into a cloud. In the hardware camp moving a little bit to the open side you have another player called Morph Labs also hardware but their plays like best performance price performance, commodity components and they're generally kind of more open so they're more aligned with the open message and naturally you have all the incumbents everyone has their own approach but they're all there and they're trying to ride the open stack wave to somehow creatively package bundle solutions that somehow tie together with open stack. Software camp it's, you know, pretty much the pros and cons of it are really just less pronounced pros and cons of the hardware camp I think. It's a little bit less of an R&D investment you can still build a leverage business model the cons is that it's somewhat hard to penetrate that market because enterprises are not used to buying, you know, pistons, they're used to buying VMware and they're not used to buying cloud they're used to buying data center automation so you have to invest a lot of money to change that mentality and it's enterprise sales process expensive and hard although you can sell through VARs you can't expect VARs to ever create demand for you they might they might help resell maybe pieces of hardware with your software bundled into it but unless everybody knows about it and wants to buy it to begin with you won't see guys like Ingram Micro for instance selling okay this is like a Dell appliance with piston running on it buy by buy so as long as somebody is asking for it they'll give it but otherwise the kind of a battle against the existing brand perceptions and you to quickly gloss over the key players so cloud scaling their angle is we have the open stack distro for really large scale deployments telcos and large scale deployments for the enterprises so large scale web scale is their angle piston VMware killer that's kind of their business model we are the enterprise open stack that will kill VMware Swift stack exclusively focused on Swift and stack ops I think they're based out of Spain fairly popular their focus up until recently has been on a kind of a really small use cases so if I have like you know two or four nodes but I want things to work really quickly and push the button that's what you get now they released the enterprise edition having amassed a good amount of popularity so it's arguably not a bad play support camp all you see is the incumbents I won't spend too much time on this one the points I wanted to underline here is that as we're moving away from the hardware and software camp we're moving into the paradigm of companies and services that are being offered that in my opinion are much better aligned with the existing open stack momentum so existing open stack momentum is very much about open and if you're tying your messaging to open stack you're saying I am an open stack appliance I am an open stack distro it has to be that you're consistent all the way and selling hardware and licensed software with the open stack message is a little bit hard but doing something like a redhead support model with open stack that I can imagine working because that is simply aligned with people's perceptions when they come to open stack they have certain perceptions and when you tell them we'll sell your support we'll sell your software license makes less sense key players like I said the incumbents most of you probably know the positioning canonical they are the linux for the clouds all of the cloud providers and all the guests in the cloud are all Ubuntu linux that's the messaging redhead naturally we own the enterprise we own the open source ecosystem we're gonna conquer the world of open stack now and Susie's very angle is linux for mission critical applications such as for example I think like 70% of SAP workloads run on Susie and a lot of other mission critical stuff so they're mission critical angle rack space is also somewhat in this camp at least until recently their messaging has been fanatical support we are the hosting company that does fanatical support but it doesn't matter if we're doing fanatical support on hosting we can also do fanatical support on supporting the software so that's kind of been the angle we're slightly moved away from this at this point but I think kind of the general messaging is still largely the same and the SAP camp you sell services some pros is that you can start monetizing immediately so as soon as you sell your first account you start making money it naturally makes full use of open stack momentum because you're not introducing any kind of products or agendas that might conflict with the existing kind of incumbent open stack messaging and the good thing also is that you get exposed to a lot of opportunities and you can learn about customers buying patterns as they relate to the open stack ecosystem as you engage in real projects the big two cons of this approach is business scales linearly there's no way to build a leveraged business model on services and this business is very short lived eventually it will get commoditized and that'll be the end of our open stack value in the context of the services play so I'd like to kind of close by sharing the Miranda story we've talked about the theory of making money on open stack we looked at how other players are trying to make money on open stack at least as you look at them from the outside in and now I can share with you the story of how we try to make money on open stack and that's kind of a story from inside out so the story is fairly simple and it goes something as follows before open stack started and before we got you know deeply involved in it about 18 months ago we were somewhat of a generic application infrastructure services company anything that's cool an application infrastructure such as Big Data or eucalyptuscloud.com Solar and Lucene we get our hands on and we try to build buckets of expertise and we're able to more or less sustain the business and make money that way when open stack came about we said ooh another cool thing how about we play with that we played with it it was bare release actually what we did is we took it and we deployed private cloud for our internal IT I mean the bare release was buggy it was horrible it was not easy to do but we did like the kind of the layered architecture approach and we also liked naturally the amount of market hype and community momentum that was there so we said okay we want to explore this a little bit further so what I did is I called Mark Collier who is now the CEO of the company and he said hey you know we kind of want to get involved and see if we can explore this further Mark was at Rackspace then he was a very busy guy so he said I don't know you Boris don't talk to me talk to Lauren sell she deals with fools like you so I talked to Lauren and Lauren said okay what you guys should do is if you want to get involved how about you make a meetup the Bay Area local meetup for open stack that is focused on kind of you know the business ecosystem and potential end users that's an opportunity for you how about you do this so we said yay let's do this and we figured we're going to do kind of a meetup of 20 people showing up got a small room we ended up with 120 people showing up and had to upgrade the room two times and shortly after the meetup some interesting things started happening kind of the ecosystem learned a little bit that there's a company that is doing services in open stack so we started literally getting inbound interest a randy bias whom some of you may know called and said oh you guys do services around open stack can you please help us out you know we have we do a lot of open stack stuff we could really use a partner and then Ray O'Brien who is CIO at NASA at a time who ran the Nebula project called us and he said hey Boris you know I see you doing open stack stuff all of my guys left to do their own startups like Nebula and Piston and I don't have anybody to support my open stack infrastructure do you guys know anything about it can you help me out so we said yes and then some other people from Nexenta that you don't know have also called us and said open stack is cool and we want to integrate with it do it for us and these are not the only three but that kind of trend so we started looking at it and I came to the rest of the company shareholders and I said that you know guys this is kind of an interesting trend how about we now focus on open stack and they looked at it and said no it's too risky it's too narrow bad idea we will be cloud services company we won't be an open stack company and I said well cloud services that doesn't make any sense who's going to buy cloud services doesn't I don't understand it and they said well yeah just because you're young and stupid so stop talking to us so what we did is we decided to call a professional marketing consultant Geva Perry whom again many of you might know and we asked them you know we're seeing some trend here that people want us for open stack what should we do at the same time we have all this legacy business we're doing all this great Hadoop stuff and big data stuff and all that stuff I mean we don't want to change our messaging we should focus on open stack should make a website that says that you're all open stack buy some Google AdWords to point to the open stack stuff you know do some more meetups and write some more blogs about open stack and it's going to be cool so I shared that sentiment and I kind of brought his recommendation to the management and they said ooh now the marketing consultant said it not you young stupid fool alright let's do it so what we did is we changed our website we bought a bunch of Google AdWords that pointed to it it was no longer cloud services it was now open stack and now we kind of grew to be the big kahuna in the open stack services so we are the one of the sponsors of the foundation we arguably I'm pretty sure that it's true because I know the ecosystem we are the largest services player in the ecosystem we have already 70 actually people that are exclusively dedicated to doing open stack projects we've done already at this point close to 30 open stack related projects you know some are for fairly notable customers such as Webex that presented a keynote this morning and we've been growing at about 100% per year since we committed to open stack so that's our story thank you and I'm open to questions if anybody wants to ask them but you don't have to if you don't