 Have a seat. Jonathan comes from the OpenStack Foundation. Probably many of you here know him. Jonathan, start by telling us where you think OpenStack is today. So, you know, OpenStack started in July of 2010, so, you know, a little over three and a half years ago. And I think that if I kind of look at the phases of OpenStack, the first phase of it was really in 2010 and 2011 forming the community, reaching out to organizations outside of Rackspace and NASA, who were the two organizations that kind of kicked it off, getting involvement, getting development, getting a process for code to sort of move through the pipeline, and really, you know, kind of setting up that piece of it. And then in 2012 we actually created a nonprofit foundation because at this point there was a lot of momentum, a lot of participants, and it was clear that it should move outside of kind of being managed by Rackspace and into something that had a broad base of support and governance. And, you know, that was, I think, a lot of what went on in 2012. And that was when I would say we started to see the first traces of adoption in a serious way in enterprises and in different organizations. And like you said, you know, a lot of these are early adopters at the time, companies like Intel who's been building an OpenStack environment out for two and a half years now and doing a lot of different workloads on it. That's when they started. And as we come into 2014, what I think I see is we are at a point where we have an extremely strong base of providers, of developers who are producing technology and a growing base of users. And I think that we're in a really key spot now where it's time to get operators and cloud end users to really start to be more embedded into these processes that we put in place. Because as you say, you know, this is the time when we start to transition from early adopters and technology companies to a broader base of adoption. And there are a lot of other requirements and a lot of other input that needs to come into it. So what do you think needs to be done to drive sort of past that initial early adoption? And how does that fit with what the foundation is doing in 2014? So an interesting thing is if you look at some of the biggest users that we have, a lot of them are working with a vendor in some way but also doing a lot of it themselves. And they tie in different components from the storage side or from the networking side. And then there are pieces that they go straight to the code base for. And I think that over the long term, you know, broad-based adoption is going to happen in concert with various other companies. And those companies are taking the bits that make up OpenStack which, you know, there's a lot of technology there in a lot of different areas and they're turning it into products. And I think that is a way that traditional enterprises are used to consuming technology. At the same time, I think that one of the things that's changing inside of enterprises is that software is becoming much more of a core piece of people's business now than it ever has been before. And, you know, we have an OpenStack user that's, they are not a technology company. They're a very large retail organization. And what they sell has nothing to do with technology. But they have stores all over the place and they tie together all of their POC systems. They have large professional customers that they deal with on a regular basis. So they have advanced CRM systems. And their software is actually very important to them in doing their business, much more so than it was five years ago. And so there are a lot of organizations where that's the case now. And so the infrastructure to enable their developers to iterate rapidly, to scale, to try out new functionality is more important to more organizations than ever before. So these vendor organizations involved in OpenStack both cooperate and compete. And as a result, we've had a lot of, for instance, different distributions, proprietary extensions to OpenStack, which ends up creating challenges for hybrid interoperability and really also the ability to switch between vendors. What is the foundation doing to address the needs of users to really understand what OpenStack is? Yeah, this is a big focus in 2014. There are a couple of initiatives that we're spending a lot of our resources at the foundation on. And the foundation staff is very small. So our goal is really to basically coordinate activities within the community and get a lot of things moving in a direction and getting people involved to help actually go carry that out. But two of the key efforts that I would say are there's a process that's going on right now that's codenamed DeafCore or core definition. And it's kind of a series of steps that have been taken over the last year to get to an automated testing framework for interoperability, for downstream products that come out of OpenStack. You know, you can take the OpenStack code and do a lot of different things with it. It's very broad technology. That's part of what makes it so powerful, but it also can make it complicated. And you can end up with an OpenStack cloud that looks one way and another OpenStack cloud that looks totally different. That's fine, you know, because people have different needs. But what we want to make sure is that it's clear where those differences are, which things are going to work in one area, which things are not going to work. And so the DeafCore initiative is something that started up in November. And by the summit in May, we're going to have automated testing. We're going to have a way for publishing these tests so that it's easy for users to be able to go in and understand different solutions and, you know, how they match up to kind of a standard OpenStack installation where the differences are. And that's one initiative. Another thing that we're doing in 2014 is a series of different user engagement events to basically pull users more directly into the process. There's room for users right now in our process, but it's open source and it can be intimidating to come jump into an open source community of 2,000 developers, you know. And so there are some smaller events that we're putting together with operators and users. And we're also doing some, you know, just old-fashioned relationship building between different users. We actually had a dinner last night with several users here in the Bay Area. You know, three companies that are some of the biggest brands in the world all using OpenStack and doing it at, you know, pretty large scale. And it was interesting to see them get together. And within, you know, 10 seconds of them sitting down at the table, they were just peppering each other with questions. Oh, you know, what do you do over here? And how did you handle this? And what did you find out when you started using this project? And oh, you're looking at this. And I think that there's an incredible power available out there if we can get those people talking, communicating, get them into the community. And so we have a number of initiatives that we want to spin up around both operators and cloud consumers, you know, end users who are building applications on top of OpenStack this year. How was the OpenStack roadmap built? And how does that, how does basically user desires factor into project priorities? Yeah, that's a really good question. Especially against the come with code. Yeah. Well, so the OpenStack roadmap is built through a six month release process. And we are about midway through the current release cycle. And so with the six month cycle, we do a release, and then we have an event. It's called our design summit. The next one is in May. And at that event, developers and users and people get together and talk about, you know, what the priority is going to be for the next release. And then they go off and they build it over the course of the next six months. One of the things that we've heard, the technical leaders love getting that feedback, but doing it at the summit in a lot of cases is too late to bring in a big change, a big requirement. And so we're actually creating mid-cycle meetups, and some of these are focused just on users, where we are working with the user committee of the foundation to come up with some prioritized lists two months ahead of time that we're going to feed to the technical leadership so that those discussions can happen ahead of time. And, you know, you mentioned come with code. It's definitely a code speaks kind of community, but, you know, every technical leader in OpenStack that I have talked to or that has interacted with a user is really excited to get that feedback because that's why we build software. It's for people to go out and use it, you know. And so what we at the foundation want to do is just facilitate those conversations at the right time in the cycle and to be able to, you know, have a good impact on the software. And so in terms of where OpenStack is going in this next year, what's going to be significant across Icehouse and Juno? So I think that, you know, there are a lot of new programs and technology components that have spun up around OpenStack, and those, you know, often get a lot of coverage and a lot of excitement. And, you know, you have things like the orchestration system and the metering and monitoring system that came in the last time around. And, you know, you have the database as a service and all these kinds of things. And then there's a whole roadmap of other things that are out there. But I think that what's really valuable that's happening right now is these users who are getting involved. We had, I think, over a dozen pretty major users that were code contributors in the last release. And what they are doing is they are driving it towards maturity and towards being much more operable, towards being upgradeable, you know, these kinds of things. And I think that that, to your points earlier, those are the kinds of things that bridge that divide between early adopters and kind of massive option. And in that context, as you look at OpenStack's future and you look at innovation in OpenStack, to what degree do you expect that innovation is going to be driven by OpenStack versus OpenStack seeking to essentially commoditize innovations from the rest of the ecosystem, whether it's Amazon or Google or Microsoft's innovations, let's say. That's an interesting question, you know. I think that if I think back to where we started, it was much more about trying to come up with a strong, viable option for some of the technology ideas that were already out there. But what we've seen as this community has grown is it's actually drawn in a lot of very strong technical talent from around the world in a lot of different subject areas. And more and more, you know, there are things that are being done in OpenStack that then get pulled into other, in some cases, into other proprietary companies, into other open source projects. And it's not just on the cloud side. One of the things that I think is really interesting that's happened is it's one of the largest and fastest growing open source projects ever. And we get tens of thousands of contributions. And in order to be able to handle that at the kind of scale that we're at with a distributed system like this, there is a lot of testing that we have to do and a lot of complicated dependencies in this testing. And so, you know, we've built an open source system for managing that. And that sounds kind of like, okay, that's interesting but not really related to cloud. But it's actually something that has been implemented inside of proprietary software development companies, inside of other open source projects, major ones. Because, you know, what really has happened is it's created this kind of center of gravity for people who want to work on hard distributed problems. And so I, you know, I think that a couple of years ago I would not necessarily have seen it as like the hotbed of innovation, but I think that it just shows you the power of a community. It's becoming more and more of that. So I think looking out to the future, I think that it'll continue to be a combination, but I think we'll definitely see some strong innovations coming out. Good. Thank you. Stay put. Ken, do you want to come up? Thank you. So Ken, you work for Selenia, which is a professional services company for OpenStack. What drives adopters to look at OpenStack professional services? Well, so, you know, I think we see a wide array of adoption out there. And I think different segments of the population come for different reasons. If you're a service provider, you're really coming probably for integration. You've probably rolled your own because you need some differentiation in your service, so you need source code control, but you do need to integrate it into things you already have there. You've got a billing system. You've got some kind of operational system and things like that. And so it's really around the technical bits of integration and things. A lot of our work, though, is actually in the enterprise. And the enterprise comes at a much different entry point. They really come in at what we call a strategy point. And most of the time when we get involved with clients in the enterprise, at least, their first thought is, do I need a cloud or how would I actually take advantage of cloud? And it's not about OpenStack. It's not about private or public clouds. It's, should I be using cloud? And if so, how would I get onto that? And there's certainly technical elements to it, but actually a lot of it's around process and governance, is if I were going to use a private cloud or a public cloud, would I make sure that I still have governance of who's running particular jobs? How do I know where my data is? How do I know my data's safe? And those kind of things. And so it really depends on what kind of situation you're coming from, of why people come to, you know, someone who's actually going to help them with consulting. So you get involved initially strategically. So you get involved in the deployment side as well? Absolutely. So, you know, I think it's really difficult to have someone come in and do generic cloud strategy for you because it's really easy to get disconnected from the realities of how hard it is to actually deploy this, whether you're moving to a public cloud or actually implementing a private cloud. And, you know, the answer tends to be as you're going to do both. And so there is a lot of technical issues in there and it's really difficult to just have someone come in and do a strategy and kind of walk away from it and think that magically someone's going to be able to pick that up and actually make that happen for you. So we do a lot of both architecture and integration and what we've really found is, is challenging as things can be on the technical side around implementing which is fairly well known. Most of the people actually have the problems on the operations side. It's, I've built it. I don't know how to get my actual business divisions to put their apps onto it. They don't know how to write for it. They don't know how to migrate to it. My operational people are used to having being, you know, effectively level zero support people and being able to go to a back line vendor which, you know, a lot of them don't have a vendor right now to do that too. And simply, you know, a lot of the tools that are out there that they're used to don't exist in the OpenStack. And not just in OpenStack, you can go to CloudStack or vCloud and have the same problem. Operating clouds are difficult. So what do people typically find to be most difficult when initially deploying OpenStack? You know, so I think there's two parts that are difficult for them. One is simply the scoping and overall project management of it. I think a lot of them have a wild idea of what they've seen Amazon Web Services be able to do and think they're going to be able to replicate eight years of 500 engineers at Amazon in their data center in three weeks. And so there's a scope problem and an overall goaling problem. And then there's technical issues. There's technical issues with any cloud out there. And I'll be honest with you, most of them probably boil down to networking. I think networking is probably the most difficult part for anyone. And it's also the part where you have to have the most legacy integration in. It's pretty easy to go out and buy all new virtualization pieces, buy all new storage. But at the end of the day, you probably have to integrate into the corporate network in some way, and that's where a lot of problems tend to happen also. Probably the least mature piece of OpenStack or any of the pieces out there. Probably also with the highest potential, though, to really change how enterprises do things. And so for an average organization, not necessarily an organization that's incredibly technically savvy, how does it take to basically do that kind of knowledge transfer where you can leave and the organization is able to take over comfortably installing, upgrading, maintaining that OpenStack-based cloud themselves? So I think it's depending on how much they're willing to actually change their organization and how much they actually need to change it. You talked earlier about some people actually wanting to take their initial landscape that they already have and then move actual workloads into the new cloud versus people who start up and do a net new. We really don't see a lot of people today taking that landscape that they have. Most of them are kind of boxing that landscape and saying let's put new apps out here on the new cloud, especially if the workloads fit for it. For those, if they're willing to come out and create a new organization, make some organizational changes and perhaps break down some of the walls that they've seen between network, virtualization, servers, data center and storage people, start getting them to talk together either in a specific organization or quite truthfully hire in for that, it can be anywhere as short as a month or two. Other places, we've talked to them for years. It's much more of a change management process than it is simply a technology transfer. As you look at where OpenStack is going over the next couple of releases, do you think that professional services are going to become more important in the ecosystem or do you think that more users are going to find it easy to be able to do all the implementation they need without the use of professional services? So I think what you'll find is it depends what professional services you're talking to. So if you talk about a life cycle where you start with strategy, architecture, you actually deploy it and then you start talking about adopting where you're actually putting workloads on it and managing it, I think where you're going to find is you have value at each end and in the middle, you're going to see a valley of death. If you're just out there to implement as a professional services organization, that is going to get easier. Every single vendor out there is writing an installer and trying to get to that magical MySQL five-minute install. We're not there yet, but over time that's going to be easier and easier and it's going to be less and less value. On the other hand, the strategy and the adoption of it, being able to actually drive monetary value on this side and being able to make sure that you have the right scope and strategic objectives on this side, I think those will become more and more valuable and especially as it comes more complex where hybrid clouds become a reality, I think either end of that becomes much more valuable. And as customers go through trying to decide what are the components that go into the solution, what are the critical decision points that you think they're having to decide on, whether it's server architecture, storage architecture, other tools? So I think they have a lot of different decisions there and it's not just they can... I think a lot of people would like to take those decisions in silos, what's the best servers? And I get this a lot, what are the best servers? Give me two servers that I can go to. Can I go do Quanta or WisDrum or can I put it on Dell? And they want to silo those decisions because that's how they've made decisions before. It's a classic purchasing decision. The problem is, is Opensack's a framework and it is a solution. Everything needs to work together. And today, and I think not only today, but in the future, not every solution, you cannot pick one thing from everything and just magically make everything work. There are solutions that work together. There are solutions that don't really work together. You can't make your decisions in isolation. You need to make a holistic decision, both on the technology side, but also on how you're actually going to aim for workloads. I think a lot of people think, well, this is the new virtualization. I will go ahead and take any virtualization workload I had before and put it on there. Yeah, some of those may work, but you may not be driving the most value out of that. You really need to look at what cloud is going to drive the most value for you and think about then how do I design my cloud for that and then how do I actually move workloads onto it? Well, you're a typical customer. What constitutes success with OpenStack? So most of our customers, we try to start small for them with a proof of concept or a pilot, like pretty much everyone does. And usually we go in in the strategy area and highlight specific workloads that we can drive specific value from. And in most of the time, those are new workloads. And increasingly, we've found that the big data analytics are the things that actually make your TCO or your actual benefits really drive the economics for that cloud. And so we'll come in and help them with driving those KPIs that they need to figure out what that would be. A lot of times there's going to be a cost component in it, definitely. But for a lot of the people that I'm looking at, they're actually getting into new things they've not been able to do before. So I've got a top five automotive company not typically a person you probably see as an early adopter. They have a huge amount of data out there that right now, in general, they tend to use Microsoft Excel on people's spreadsheets. And it takes everything from telematics data to warranty data to data actually in sensors on race tracks when they go out and test these cars. And now they're able to actually centralize all of that from all of their different areas and even with people out driving in the field, they're able to do things they haven't been able to do before. So it's not just an economic, they're actually breaking in and doing new things with it, which, you know, in general, it's hard to express sometimes in economic terms. In about five minutes I'm going to take questions from the audience. So if you've got questions, begin thinking about them. A question for both of you. What do you think is going to be most important to broaden the appeal of open stack beyond early adopters? So I'll answer that. So I think today a lot of people look for open stack and are looking for a product, more of a product as they have imagined it before. And I think today you see kind of products on two ends from the vendors. One is I've taken the bits and really just packaged them and I'm going to give you support behind it, but really nothing more than that. It's a pretty light packaging and it's probably thrown into your distro. On the other end you have a very tightly cast solution all the way down to telling you what hardware it is. I think people want something in between there, which is they want all of the pretty installers, the great looking GUIs, they want some management tools around it, things that you would normally see in perhaps like an ERP system. They want to be able to see the ecosystem that's there. They want a true product and I think what we're going to start seeing, and we already see that with some of the large vendors today, we're starting to converge towards that middle and I think towards the end of this year or early next year, we're going to start seeing that. And that's the only thing that's really, in my opinion, has cut cloud stack in any of the discussions today. Let's face it, they've got a great looking GUI. Past that, things kind of fall apart on you, but they've got a great looking GUI. I think as we start getting that polish on some of the areas, especially in the products that are supported, I think that's going to start disappearing. I think that that's... I'm not going to disagree with anything Ken said there. I think that one other thing that you actually touched on, I was actually at a Gardner conference in December. And I was talking with different people and they were talking about OpenStack and their plans for it. And then one of these guys I was talking to is that he's at a very large enterprise and he said, I want to do OpenStack. Who do I hire? How do I build out that team? And he's like, this cloud thing is very different from how we've run technology before. It's not like I need you to point me to an individual. Tell me what job I hire for. So I think that that's still one of the things you mentioned earlier in your introduction that we're on the front edge of private cloud adoption. And there are companies out there who have a business need that's driving them there faster. And I don't think it's just broken down by technology or non-technology company. It's really opportunity-based. But as these other companies come along, they're having to kind of... And sometimes it's much more... It's much more trouble on the organizational front than the technology front for them. And so I think that's part of it as well is just coming up with patterns about how you build out a cloud ops team, how you structure your organization to provide technology services rather than servers for technology. And some of those kinds of things are, I think, the key pieces on the organizational side that hold back adoption of OpenStack and cloud in general. So what is the skill set of the ideal sort of OpenStack administrator OpenStack champion in an organization? Yeah, it's all over the map. It's interesting because one of the things that we started last year was a training program within the foundation to help aggregate training around the world. And we have... You can go to openstack.org slash training and find courses in something like 40 or 50 different cities around the world. There's a lot of training out there that's available. And as we go through that process and talk to people, they come from all sorts of different parts of the organization. In some cases, they're developers who couldn't use a public cloud. You know, they... For whatever business requirements, but they want that flexibility. In some cases, it's a... It's, you know, a DevOps kind of organization anyway, and so it makes sense for them to move this direction. And in some cases, it's coming from the architecture and engineering staff that's running the infrastructure right now, and they're getting a lot of inbound demand for this kind of functionality. So it really depends on how it's... If it's kind of being pushed as an organizational thing or if it's a poll from different consumers within the organization. And I think that it just... Again, that speaks to kind of the early stage of it all. There's not a... I would say that I haven't seen one standard process for how OpenStack finds its way into an org. So what skills does someone need to administer OpenStack at this stage in its maturity? Yeah, it's the organizations that I would say have been most successful have strong systems engineers and who are very comfortable in kind of looking at the internals of what they're running. You know, I think as Ken said, there are a number of products that are coming online now that are everything from distributions of the software to appliances and those smooth out some of those rough edges and make it something that's more consumable. And I think that's... We're just getting to that stage where an administrator can operate an environment like this given the right configuration. I think the other thing that you see out there is you need a network department that actually understands applications. And that can be either the biggest roadblock you have or your biggest benefit as you're putting this in. If the network people really understand applications and enabling applications, they can really smooth all of this over for you. And so I think that's a key part. And then on the adoption side, I mean, it's enough to put it in and operate it, but eventually your application people need to be able to write applications for it. And being able to go into that side of the organization also and make sure that they understand how they might be able to use this either in a test dev or a production way, we spend a lot of time over there also to make sure that they actually monetize this later. Questions from the audience? Or a new customer. So the question there was what are the use cases that you're seeing out there and then what are our favorite use cases that are out there? So I see two of them that come up consistently and this depends on who's actually championing it within the organization for you. But consistently you will see analytics or big data, you will see some kind of mobility, whether it's mobile phones or cars or something like that, and the ability to actually project applications for them. So those two come up fairly consistently and especially if you have a CTO or a Chief Innovation Officer who's leading the charge, that tends to be that new application there. However, there are other places especially if you're a SaaS provider today, it's their SaaS application and they're wanting to, in general, repatriate some amount of their real estate back from AWS or someone else and bring that in-house for them. And so those are kind of the top three that I see out there today. I love the big data one myself. It drives a lot of the real key, I think, benefits of having cloud. It's also just one of my interests because it hits all of the areas there and kind of stresses any cloud that you have and actually lets you use some of the really interesting features of OpenStack that are out there. But other people will have their preferences also. I think that a lot of it is applications that run well in kind of a traditional three-tier application architecture. That's what a lot of people are deploying. In terms of things that I think are interesting, I actually have found it interesting to see some of the... They're sort of sassy, but it's the applications behind appliances like the Comcast Xfinity use case or the Sony PlayStation use case where the thing on the other end is not someone in a browser or someone sitting at a computer. They're using something that you... It actually puts a lot of load on infrastructure somewhere back behind the scenes that's powered by OpenStack, but you would never think of OpenStack as being a component of that, and I think those are always interesting. I think that leads to where you might see the next wave of big applications, the Internet of Things, where it's a thousand-ness thermometer or thermostats that are out there. I think that might be the thing that drives that kind of next wave that you see. Stand up if you've got a question and someone will come to you with a mic. Stay standing if you've got a question, actually. Thank you. For net new applications, what are the reasons that you're seeing enterprises choose OpenStack over a public cloud like Amazon or GCE? Is it, you know, fear of losing the jobs or is it regulatory compliance reasons or cost or what do you see? Yeah, I mean, so for most people today where I've seen them, especially in big data, move to a private cloud is, A, simply they know their security stance will not allow them to do that, whether it's regulatory or compliance, or it's simply that data is too proprietary for them. However, there are a large number of people that probably don't have those rules, but they don't actually have the governance in place to know if that data is actually proprietary or not. So I have more than a few clients where they just haven't gone through and looked and said, this is proprietary data and I can only use it in-house versus this is data that could go out-house, sorry, outside of house. And they really haven't gone through that exercise yet. They don't have the corporate rules around governance of what I can do outside or inside. They may be in a semi-regulated place where they have some customer proprietary data and they don't want to co-mingulate, for example, but we see a host of different reasons that people do that and most of them will default is, if I don't clearly know that I can do it, I probably will do it inside. You also see the point of because it's a batch, it's a fairly constant load and because it's a constant load, you kind of let's own the base and rent the peak and so they'll go ahead and own the base there. Someone somewhere right now is actually setting up a stack forage project for OpenStack Out-House. Thank you. All right, another question. Yeah, as OpenStack becomes more mainstream, how do you see people backing up OpenStack-based solutions as there are more focused projects in OpenStack like that's going to focus on backup recovery or what do you see people doing there? Yeah, so I'll be honest, I have not seen anyone, I haven't seen any commonality in how they're doing either backup or disaster recovery around those infrastructures. I think we're starting to see some interesting products that people are working on. I've heard about fault-tolerant versions of these of coming out. I think there's a lot of people looking at archiving solutions and such, but for that infrastructure itself, you know, I think we're waiting for, that's hopefully a feature of products that are coming to market that will have a backup. Today, almost all of that is some kind of bespoke thing that we integrate for that particular client. And, you know, depending on how large the cloud is, it may be something where we're backing up one cloud to the other, one site to another. It might be one where we're taking and backing up particular parts of it into their legacy landscape. But today, I haven't seen any commonality or best practice probably there. Another question? Yeah, right here, to your right. Is there any use case that just could not be done any other way? I'm looking for sort of indisputable advantages of OpenStack. So, in general, I wouldn't say could not be done any other way, but could not be done economically any other way. A lot of the big data fits into that. I mean, I come into a lot of the engagements thinking that it's going to be an application and like when I went to our automobile client, I was sure it was, we were going to be projecting interesting apps to the dashboard and cool, like we were doing maps and all these things. But when you went down to, you know, the numbers and what was going to prove the TCO for that cloud, it was, wow, we've got a lot of data and putting that in a public cloud or in a legacy, what they, well, the Teradata, it was a no-brainer to go to cloud for that and it was, didn't even have to have a conversation about it. So, in general, you know, I hate to keep harping on it, but it tends to be that data element. Well, it's computer science, so there are always an infinite number of ways to do everything. But I think that, you know, the, that, you know, we have some other users that are going to be speaking today and it's a good question for them for, you know, I think, you know, the points that led them to this. The thing that I think I've heard over and over and over again is it's about allowing them to move more quickly and, you know, yes, there are other ways to accomplish agility inside of IT, but, you know, OpenStack lets them do it in a way that is flexible on the deployment side, that lets them work with technology or vendors that they already know, lets them customize it to their needs, it lets them put it in-house in their security posture and when you add up all of those things, then it just makes sense for these people to, you know, to go this path. And so how do users adapt their security and regulatory compliance needs for OpenStack? It's, that's another one of those things that is that it's different for, you know, a lot of the different use cases. The biggest piece of it in many cases is just being able to get it inside of their environment, inside of their network, you know, that's kind of step one. And then after that, it depends on how strict those requirements are and the kinds of things that they need to do. But, you know, it's tying it into their existing intrusion detection systems and, you know, some of this was some of the key drivers behind Neutron in the beginning to set up a framework for being able to have a more richly manageable network topology inside of your virtualized environment. And then, you know, other pieces of it tie in. There's always, you know, layers with security and you can go down to the extensions and the processor to verify the hardware and what's on this hardware. So it really varies all over the map. The interesting thing is that, you know, you can go to a lot of the different vendors that work inside of OpenStack and they have, they expose those things through OpenStack so you can take advantage of it where you need it. OpenStack actually is very rich in the points of governance that provides you. Not all the tools are great. I mean, you're in JSON for some things and perhaps it's not a nice GUI for it. But it does provide the points of governance both in Keystone as well as in your service catalog within Glantz in different areas to actually provide the similar points of governance that you may have in your legacy. It's really about, though, talking about how you're going to read you governance and move from a deny everything, to come to a person for an exception to go do something to a whitelist approach where this person has the ability to do these things. If you go outside of that, then you perhaps need some kind of human intervention. And we've had this problem, you know, other places where we had a client, for example, here in the States, a financial company, it took them somewhere around 85 to 95 days to put a new server on the floor. They went to virtualization and that was 75 days to put a new BM on the floor and they went to cloud and it was only 73 days to put a new cloud instance there. So obviously that's not a technology problem. That is a process problem and that's a lot of what this governance and security is really about. And it's been able to have the discussions about how do I do this in a sane and rational way without exposing, hopefully, too much risk to the corporation. Yeah. I wonder if you could comment on your third use case, the re-patriotization of public cloud applications. What's the dynamics involved there, the business cases and how did those work out? So I've heard, depending on which SaaS provider you're at, but almost everyone that I've talked to, they have a number for Amazon where they believe that they get above that number of a dollar spend at Amazon, that they believe they're getting diminishing returns on their service and everyone has that number and I've heard wildly different numbers about what it is but everyone knows what that number is and it appears today and when they start getting towards that number they're looking about how can I move my steady state parts of my application back on to premises and be able to do, you know, be able to service that locally, well then I actually just rent the spike or the peak out there. Now having said that, that doesn't mean that there's magic around creating a hybrid cloud that automatically auto-bursts in AWS and things. I think though everyone's looking at that today about how can I actually do that, how can I partition or architect my application to be able to do that, where it might be something as simple as, you know, I'll do batch on premise at night or during the day on site because that's a fairly relatively constant workload and by the way I have my data there and then I'll just run the online part which happens to be much more spiky and rent that part but I think it's an evolving one right now it's just one that I'm seeing. I can't necessarily say I have the perfect answer because I think it's going to be very it'll be very different depending on your application. Thank you both. We've got one more question on the left-hand side or speakers right. So a lot of what's been talked about is organizational adoption of OpenStack for private clouds. Can you give us a little bit of a view of OpenStack-based public cloud platforms apart from Rackspace and who's moving up and what are some of the choices out there today that are commercially available? So, yeah, the public clouds that are running on OpenStack, you know, Rackspace is obviously one that people know about. HP Cloud, they're here today. I think Bill is going to be speaking later, so Internet was also here today? Yeah, and Internet is here today as well. You know, there are a number of other companies out there in the hosting and cloud space like Bluehost and Dreamhost and you know Vance and CloudWatt over in France. There are companies in Asia and all over the world. This is actually getting back to some of the points we were talking about earlier. It's a common question and so one of the things that we want to do this year in the foundation is really organize and aggregate the different pieces of the OpenStack ecosystem in a way where if you're looking for an appliance, if you're looking for a public cloud, you're able to go navigate that more easily and you know, answer that question very clearly by region, by capability. Again with the DEF core process and the interoperability you're able to tell my application needs this set of APIs, they support them. So they're a lot out there but it can be kind of hard to find that and I think that's one of the things that we want to fix this year. Thanks. Alright, thank you. Great. Thank you very much. We're going to bring up our next panel