 Hello, everybody. My name is Azmer Mohamed. I run products at Cloud Scaling. We're a startup based out of San Francisco. And today, I want to talk about hybrid clouds and how we're building hybrid clouds using the Open Slack. So I think these are the things that a lot of our customers are wondering around when they're either in the public cloud and they want to bring their applications home. What are the things that they have to really think about? And we spent a lot of time really making sure that you have a solution that's built around Open Slack. But it's a rethink of your entire stack from the certified hardware components to the reference to architecture and obviously the software that you run in order to build your modern cloud. And for Cloud Scaling, what we focus on relative to our version of Open Slack is around production readiness. So we have customers that want to deploy clouds today. And these are clouds that span three to five racks of gear. And also, public cloud federation. So a lot of our customers want to have a hybrid cloud strategy. And they want to be able to federate out to a public cloud, like an Amazon Web Services or a Google Compute Engine. But to be honest with you, we're agnostic to the type of public cloud that you would use. What we want to focus on is to help you build your private elastic cloud. So that's a key focus. But as I talked about this today, this really the notion of a private or public cloud really is not different from implementation standpoint. It's just easier for me to go show you what that looks like. So those are the slides. And I'll try to run through a live demo here and really show the power of Open Slack and also building a hybrid cloud strategy around it. Everybody can see that? OK. So this is a standard demo that we do at Cloud Scaling for our customers to really show what it means when we talk about a hybrid cloud. And what's really important to us is to build a cloud that has API compatibility but also behavioral compatibility. Let me show you what that looks like. So I've built a bunch of scripts inside here. So I'm going to use a tool. Actually, it's a tool from Eucalyptus. It's a great tool for us to use in terms of running the same commands against Amazon Web Services and also against our demo cloud that we have in San Francisco. So this scrolled off here, I apologize. So I'm going to go ahead and, first of all, and show you what these. So I built a bunch of scripts. And all they are, they're issuing commands. So I've got a set of commands to source my credentials to EC2. If I were to go, there's another one that I use for OCS that connects it to my cloud. And then I've got a bunch of commands to create key pairs. And so what I'm going to do is I'm going to issue a set of commands to Amazon Web Services. I'm going to issue the exact same commands against OCS Go, which is our demo cloud. And I'll show you the last thing I'll show you today is OCS Go can actually sign up and get a 30-day free trial on our demo cloud. So what I'm going to do first, I'm going to go ahead and connect up to EC2. Maybe I can zoom out. Can you guys see that? Yeah, this is not going to work. Let me change the resolution here, guys, sorry. So as you can see over here, I'm connected to Amazon Web Services on the West Coast. So I'm going to go ahead and create my key pair. I'm going to create a security group. And I'm going to go create instances. And so as you can see here, I'm putting in EC2 on this. And the only thing I'm going to do is I'm going to change this to OCS. OCS is the name of our product. And we're going to go through the exact same command set to spin up instances on an open stack cloud using the EC2 command set. So I'm going to spin up instances here on EC2. And I'm going to be lazy. All I'm going to do is I'm going to hit the up arrow. I'm going to change EC2 to OCS. And you'll see that same fidelity go through and run on an open stack cloud. So we're going to go through and source my credentials. I'm going to change this to OCS. So I'm connected to OCS Go, which is my demo cloud. I'm going to go ahead and create my key pair and change this to OCS. I'm going to go ahead and create my security group. But before I spin up instances here, I'm going to show probably a UI that you're familiar with, which is Horizon. So this is Horizon. And right now, if I go under Security, you'll see that the web tier that we just created is already there. You'll see that the key pair OCS is there. But there are no instances at this point. There are no instances. All right. So we're going to go ahead and just click down here. I'm going to go ahead and create instances OCS. And we'll start seeing instances come up here. There you go. So they're spawning right now, right? So what you see is, on one hand, I'm issuing commands that are EC2 commands that you have on AWS. But it's running against a private cloud that's based around open stack. And that's really the power of open stack that we see. It's open and agnostic. You've got all these API sets that you can run again. So you can obviously can run your Nova commands. And that's fully supported. But you can also run AWS and also Google Compute Engine commands against the same API. All right. So that's great. So I've only showed you a CLI command. And I've shown you Horizon. What about when you want to do multi-cloud management? So I'm going to do a short demo here of a multi-cloud solution where partners with Whitescale and Whitescale is well known in the Amazon community in terms of managing a variety of loads on their public cloud. But what they've also done is they've also integrated into open stack. And that's what we see here. If you go under Settings and we go under Clouds, you'll see that in addition to AWS and all the regions that are supported, you also have open stack that's supported clouds. You can actually have a hybrid cloud strategy implement that's used in Whitescale. You have your public cloud on one side. And you can have your private cloud based on open stack on the other side. Well, that's not where the fund ends. You can go ahead now and be able to go into the dashboard here. You can actually, the power of this is really less about the fact that you can do a hybrid strategy. But you can deploy an application across both your private and public clouds. And so what you see over here down here is, so I've actually got a couple of apps that are deployed that can span across both a private and public cloud. And these apps that people are building on these new elastic clouds are different. Because in addition, who's familiar with VMware? No, nobody's familiar with VMware? So one of the more popular features on vSphere is VMware HA. And what it does is allows you to protect against failure of your server hardware. But a lot of customers that are building applications for open stack are building it so that the application takes care of uptime. And so what I'm going to show you in right scale is the ability for you to write an application that runs on an open stack cloud that has no HA, but it can restart itself if there's a failure in the application. So I'm going to go ahead and show you what the deployment looks like. And the deployment is made of a memcache server. And let me show you what that looks like. So I've created a policy here that says, there needs to always be two of these servers running all the time. And if there's a failure in one, I've got to be able to spin up an instance immediately. So what I'm going to do, I've created a script here. And by the way, if you notice, these are the virtual machines that we spun up using the EC2 commands, right scale sees it. So all the things that we're doing around the cloud, there are ways, because this cloud is API driven, as long as your tool can read into the API, you're good to go. Everything is seen, so you can choose whether you want to use the tools, you want to use Horizon, you want to use right scale, or Scalar, or any other homegrown tool. It doesn't matter with open stack. So what I'm going to do here, if you notice, I've got a script to terminate one of the instances in memcache. And if I can go over to the right side here, hopefully, see here, I get to the right side, I'm going to go back to my standard display. What's going on here? OK, better. So I'm going to run that script. And what it's going to do is going to kill one of my memcache servers, and you'll see it pop up again. And we'll see it happen on a couple of different places. So I'm going to go ahead, and I'm going to run it on the oldest instance. If we go back to Horizon, and I'm going to go into Admin, you'll see that I have two instances running of memcache. So I've got a, let's see here, where is this? There's a memcache server 9, and there's also a memcache server 8 somewhere in here. Yep, that one here. So I'm going to kill memcache server 8. What you'll see after this demo, you'll see a memcache server 10 show up. So we'll go ahead and run on oldest. So we'll see some updates here on this side in a minute. Back to here. Takes a couple minutes to boot up. But this is the thing that we're seeing with customers, is that instead of building an application that is highly dependent on the infrastructure, they're allowing the application and the software to take advantage of managing the uptime. And so what you can do is you can deploy much lower cost hardware, but still deliver the same SLAs for your application. And as a developer, that's important because they have full control over the uptime of their applications. Let's see here what's going on. I'll just run this again. All right, so I've got one more minute. I may get cut off here before the script runs. But this is sort of the power of the hybrid cloud. You can actually run these tiers of memcache. You can run it in your private cloud. You can decide you want to run it on the public cloud. You can build it out. And you can autoscale a variety of things. The right scale is one of the various tools available for you to do this. Obviously, for those of you familiar with OpenStack Heat, this is effectively the same type of promise that heat brings to the table. And so we're extremely motivated to go ensure that these kind of hybrid cloud deployments across private and public continue to happen and really drive towards consumption of these new type of applications that take advantage of the resource across. But it begins with fidelity at the API level and also at the behavior level because what you want to do when you write these applications, you want them to look at the cloud and see the same basic fabric. So there's no difference. There's no translation that needs to occur. And that's where the power of OpenStack comes in because it's so open. You have all this variety of capabilities that you can have. It doesn't have to appear towards another OpenStack cloud. It can be appeared to any cloud out there and we just need to maintain the API fidelity. So it doesn't look like it ran here. I apologize. I don't know what went wrong here but you can come by the booth, cloud scaling word towards the lunch area and I'm sure it'll run again but this is something that we put a lot of promise in and something that I think you can get a lot of benefit from. So thanks again. Appreciate the time.