 Good afternoon, you hear me. Perfect, then I can start. My name is Patrick Lind, and I'm head of the product line infrastructure service and platform services service at Ericsson. For those of you that don't know what Ericsson is, I will start with three quick facts. Ericsson är den 5th largest software vendor i the world. That means that software is extremely important for us. Both the software itself, but also the platform it is running upon. The second fact, Ericsson is the world leader in mobile infrastructure. We have over 40% market share of the mobile infrastructure in the world. And this is then deployed in 180 countries all around the world. So this is then a huge network of interworking nodes that is delivering a service 24 by 7 that is mission critical. The third fact, we are, as Ericsson, extremely committed to cloud. We have a little bit less than a year ago established a business unit that's called cloud and IP. That's one of our four businesses within Ericsson. So cloud is important and open stack is also very important. A few years back we took a decision that we would be building infrastructure based on open stack, as well as making our applications run on an open stack infrastructure. We also became gold member within the open stack community a little bit more than two years ago. So this is, of course, very important for this event. So over to you, Derek. Hi, I'm Derek Collison. I'm the founder and CEO of a company called AppSero based out of San Francisco. And we deliver a trusted hybrid cloud operating system which will allow you to expand your investment in open stack seamlessly to Amazon, Google, IBM, software, and Azure. Very good. So then I will tell you why is this important for Ericsson. First I will give you a market scenario. I mean, we hear so much during this conference about DevOps. I mean the speed, the automation. But then also of course you need to manage that with the compliance. You need to manage the speed with the risk. And if you combine these two things with the third important thing to getting this totally elastic working in a hybrid or a multi cloud concept then you have captured what we think are three very important market scenarios here. And in order to be able to do this, we think that a hybrid cloud OS is the future because what we're seeing here is that open stack will be very important but it will be one of the clouds. And it's extremely important that that one can work together with all other clouds. I will now do one example of a market scenario or market opportunity that we are seeing that is relevant for us because it's then addressing our main customer segments, the telecom operators. The telecom operators, they see as a great opportunity to be able to provide cloud services to their customers, the enterprises. They want to have the possibility to provide a solution where you can deploy, orchestrate and govern diverse workloads across the clouds. And that can be anything from the enterprise on premise cloud or the operators on cloud that they have built to provide the cloud services from their data center or it can be within the public cloud context. And it's important here that you have a hybrid cloud OS to create the flexibility to use the resources where you want and need them. Something that you might not know but that is based from the experience of working with 180 countries like Ericsson is doing is that data sovereignty is going to become very important in a scenario like this. Because in many countries you have the need to guarantee that the data is staying inside the country border. And you have legislation saying that and you also need to have a trusted provider. In this case the telecom operators, our main customers are extremely equipped to provide this. So they can then provide this solution towards the enterprises where they both provide the possibility to use their existing cloud the cloud that the operators has built inside the country or a public cloud. And that's why we're so excited working together with Apsara because we think that they, through the hybrid cloud OS and the possibility through policy and governance control where the data and the applications are have the perfect solution. So that's why we want to work with you guys and now we're turning over to the more exciting things. What are you guys doing? So we're going to give a little bit of a demo. For those that saw the keynote this morning you know hybrid, true hybrid is hard. It's hard to kind of get right. We were founded about three years ago trying to solve this problem and there's a lot of very, very hard problems to solve to make this totally transparent and seamless to our users and our customers. And so as you can see here we tie together private clouds, open stack, v-sphere, bare metal to anybody that's in our public cloud ecosystem which today is Amazon, Google, IBM software and we're going to add Azure by the end of the year. And so this is immediately available to all of our customers and we're going to show you kind of what that looks like. And again, we saw this morning it's hard to actually do this. It's not a weekend project to get it to run correctly but we're going to show you what we have. And so this is our operator dashboard. It's the cluster and so Josh is going to scroll down a little bit and right now at the very top that donut is Morantis Express. This morning we announced a partnership with Morantis, a close partnership of delivering AppSera's H-cost directly on top of Morantis. But as he's scrolling you also see that every single region in Amazon is also represented as well as every single region in Google. And so this cluster that's set up it's running off of or being accessed through Josh's laptop but is being housed in both Morantis Express and then AWS and Google is what we're going to do and target some workloads. So Josh is going to deploy a simple application. There's not a lot of rocket science to this web application. It's a to-do app. And it kind of comes up and says oh these are all of the things that we are going to govern by policy for you whether from the developer standpoint or from compliance ops, net ops, sec ops within your organization. So the system is cranking this thing up. He put what's called a hard tag meaning he has via policy the ability to hard tag something to say I want this to run in AWS. So even our distributed scheduling algorithm which is again overlaid over top of all these different cloud providers is policy aware. And so now we have a to-do app which is not very exciting, it doesn't do anything, it doesn't even have persistence at this point. So now what Josh is going to do is he's going to say okay we'll bind the app and give it a postgres database. And again this is all happening in AWS. So now the app will get restarted, it binds to the database which is being backed by RDS. So it's being exposed into the platform but it's actually an RDS database, right? So now when he goes back and he adds some to-do list items they will stick, right? They'll actually be stored. And so as he's refreshing and it's there. So again all he did was he said I want this application to run and by the way this can be a Docker image which we natively support can be all different types of workloads but now what he's going to show is what happens if we don't want the developer to be able to do these types of things. And so we're going to go back to the console and show the policy system and these policy systems actually hopefully read similar to like English but if we're deploying a job into this namespace the tag should be OpenStack now. So essentially let's say I'm playing the role of compliance operator and saying hey we want these things to run on-premise in OpenStack. Now when he says apply this it says we know that this is out of policy that it's not compliant anymore. So he's actually going to go ahead and resolve that and restart the application. Now what's really interesting that's happening here is that when he started the scheduling algorithm is policy aware so it knows it has to deploy it on OpenStack now we have a full mesh network that's being rerouted and rewired on the fly both from layer 7 how his browser finds the app and how the application finds the database which is still in Amazon. The database hasn't moved and of course speed of light problems you might not want to do that but for what we're trying to show you we think it's really important to understand the network fabric all being rewired across a true hybrid cloud like this is we think it was a key to driving value and extension for OpenStack investments. So is it up? So yeah it took us 3 years but I promise you it will work every single time. So this application is being routed through layer 7 from the browser to an app in OpenStack running on Mirantis that's connected seamlessly and securely to a database inside of Amazon. So we have one more other app we want to show you. This is an application that displays a map and all it's showing is that whenever he hits refresh it's showing who which workload actually responded and where it's actually located. Now remember when we started out we showed you the cluster in totality of every region in Amazon every region in Google and of course OpenStack Mirantis and so now I'm going to ask Josh go ahead and update this thing and scale it up to 50 instances. I think it's supposed to be map and so now what he's doing is he's simply saying without any policy attached which we just had on that to do app saying you have to run an OpenStack we're saying do whatever you want with a true hybrid cloud operating system and so 50 instances are being scheduled anywhere they need to be that are best suited for what they're trying to do so now as he starts refreshing we should start seeing this thing start to pingpong on Amazon that one's from Google and San Jose that one I don't know which one it is but it's up in here I know where it is geographically I don't know what provider so hopefully you guys have a sense of what true hybrid actually looks like where you can take your OpenStack investment and seamlessly and transparently extended to any public cloud provider with compliance, security, trust which are usually bad words so thank you and we're happy to take any questions I think we've got a few minutes correct any questions I promise you it's real you can see it it really does work so the question is what kind of APIs do you have and how hard is it to port to different cloud OSes, is that correct so the way AppSera's age cost is put together is there's a routing plane which is how things are finding each other there's a management plane and then there's something called an isolation context which doesn't mean anything except for the fact that it's the notion of container management and the networks stack we have to control both to enforce policy and it's a big deal that a policy says hey you shouldn't do that but you still can easily do that that's not trustworthy so to your point exactly that runtime piece which is the smallest workpiece, that's all we have to port to the different cloud OSes any other questions? yes, so the question is do you have a full fledged API that's written on top of that and you actually also ask that so yeah, so it's restful JSON payloads open you can see a lot of it through our documentation which is public and we've already had several partners extend for example our console and our tooling to do this on to a system and then orchestrating those should be as friction free as possible so we don't want that to be overly on top of what the developers trying to do so it could be that we just present the developer with operating systems that automatically have service access points and connectivity to databases set up no matter where you go but the API is open, it's HTTP JSON it's open source you can extend it any other questions? yes so the question is how hard is it to add a new service and I'm sure that the people here through either OpenStack or other potential platform as a service providers it's always the notion that you can't actually connect anything unless you actually have it integrated in with our system if you have an IP address you can natively connect directly from our system to anything that has an IP in a port step number two is if you want us to actually present it in a different way through what we call service gateway they're trivial to write, we have open source ones you can copy, paste, do whatever you want and then the last piece is a little bit on the geek side but the power of these systems in my opinion not our system but what you guys as customers run on top of it it's not only about how the system actually connects and forms an ontology but it's how the systems are actually talking and one of the kind of hidden gems in this hybrid cloud operating system we're actually semantically aware of those and we actually present a policy driven interface to actually be notified and control those communications, totally transparent so we won't show it here but the demo that he just did our we call it semantic pipeline technology is actually in between the to-do app and the database and if you wanted to we could actually say we want to be notified anytime someone does a delete or an insert moreover we can actually say next hour or we want a custom piece of code to process this and decide what to do with it so that's usually step 3 and that's the hardest to do but we have a generic framework and you plug in the protocol engine and such any other questions? we'll stick around for a while thank you very much, I appreciate it