 Go over the direct integrations that we have with OpenStack and explain you why a Nutanix-powered cloud with OpenStack is a very powerful combination in your data center. And we will also explain the architecture of the drivers that we have built, including a demonstration given by Ashish. So very high level, what we have done with Nutanix is we have built an operating system that runs in your data center on premise. And as soon as you have deployed that operating system, you will get all the benefits that you can get from a public cloud service like AWS or GCP or Azure. So things like continuous innovation, rapid time to market, only pay for resources in your data center that you actually use. All that stuff will be delivered to you through that OS called the Nutanix Enterprise Cloud Operating System. Now, if you zoom in on the Enterprise Cloud OS, you will see that we have two big pieces, two big software buckets. The first bucket is the engine that drives everything in your data center, which is called Acropolis, marked green in this slide. And within Acropolis, we offer you all the data center services to run not just your virtualized applications, but also your dockerized apps. And you can connect your bare metal non-virtualized workloads towards the Nutanix fabric so that you can run your tier one applications like SAP or SQL that you might do not want to virtualize, also on all the goodness that Nutanix provides. Within the engine, you will get an enterprise storage class fabric called Distributed Storage Service or Distributed Storage Fabric. And if you can name any enterprise class storage feature that you have today in your data center, that is built inside Acropolis. So think about things like data locality, data tiering if you have hybrid nodes, compression, deduplication, data recovery, and backup. All that stuff is built inside Distributed Storage Service. We also offer networking services as well as security services, all built inside that Acropolis piece. The nice thing about Nutanix is that it's an OS and we support multiple types of x86 service. We have our own branded Nutanix appliances. That's how we call it. But you can also run the software on two of our OEM partners, Dell Lenovo or also on Cisco UCS service. And we recently announced a support for HPE service as well. Now, in the blue piece, that is what we call Prism. And within Prism, you can find all the things that you need to manage your data center on a day-to-day basis. So if you think about how can you expand your OpenStack cluster with a single click or with a single API call, that is part of Prism. If you want to perform things like data center capacity planning, what do I need to add in my data center to make sure that I do not hit an issue in a few weeks from now when the trend continues like it was in the past? All that stuff is built in a very intuitive matter inside Prism. Now, where does the software live? The software of Nutanix lives inside a special virtual machine on one of the many supported hypervisors. And within that virtual machine, you will find a distributed application. That distributed application is something that you will get as soon as you spin up three of what we call Nutanix nodes. So three CVMs will create a distributed application. And on top of that application, you will get virtualization features, storage features, management features. Now, within the product, we also made sure that you can move your runtimes from hypervisor A to hypervisor B. We do not want to lock yourself in into a specific hypervisor. So with a single click, you can move from, for example, ESXi hypervisor towards the embedded virtualization feature within Nutanix called AHV. Again, a single click, and you switch the hypervisor in your data center. And if you want, you can also move data from your on-premise infrastructure towards one of the public cloud services. Now, I already mentioned distributed application. That is the core engine of Nutanix. The true intelligence of Nutanix is built inside a distributed service, which is built on top of some very powerful open source components like Cassandra and Zookeeper. What we have done is we have highly tweaked those services. And we have made sure that we can use those distributed applications to offer you data center services like I already mentioned. So the architecture of Nutanix itself is extremely complicated. It's a very, very impressive architecture. But this is what we made invisible to the end customers or to yourself. Everything that you will do is driven by either API or true prism, very intuitive management interface. Some of our customers use the Nutanix platform for a variety of workloads, going from end user computing like Citrix and desktop center, but also entire private cloud deployments built on open stack, running on the Nutanix platform, as well as things like SAP via unified communication stacks can all run natively on the Nutanix platform. Now, I already mentioned it a few slides ago that within the product, you have an embedded virtualization feature called AHV. For us at Nutanix, a hypervisor should not be a separate product. It is a feature. Just like you don't think about what type of hypervisor is running within AWS. You don't care. As long as you have the functions and the enterprise features made available to you, the hypervisor should be of no care to you. And that is why we have built on top of that distributed service, that distributed application, a virtualization stack called Prism. Again, part of the product, no installations or whatsoever required. Now, changing gears a little bit. Why have we built an open stack integration? It is because we have listened to our customers. And we see a huge momentum of open stack in the enterprise. But these customers who want to deploy open stack or who have open stack upon running in a data center, they want a simple, easy to use, almost invisible stack so that they can focus on the orchestration automation in general to open stack pieces. So instead of a complicated three tier architecture, we provide a simple stack that natively integrates with open stack. How does it integrate? Very easy. We have developed four drivers for open stack, four drivers for the call services within the open stack framework, Nova, Cinder, Glance, and Neutron. And these drivers can be downloaded for free from the Nutanix portal. And because these drivers are extremely lightweight and stateless, we can scale together with you in your data center. Now, the way you have to install these drivers is we have a control plane CLI, three commands to configure the drivers, to connect the drivers towards your open stack controller, and to connect a Nutanix cluster towards the drivers. Only three commands. So in five minutes, you can connect your entire platform towards open stack. Two different deployment modes for the drivers. One, inside your controllers if you want. Now what is visualized here on the screen is that when you go to the Nutanix support portal, you can download a virtual machine. You spin on that virtual machine and the drivers will live inside that VM. So you don't have to do anything from an installation point of view inside your standalone open stack controller. Now, if you deploy these drivers inside the VM, our customers do not want that these drivers or virtual machine become the single point of failure between open stack and Nutanix. But Nutanix is by nature a highly available resilient platform. But if you have one virtual machine in between, that is a translation layer between open stack and Nutanix, then that is a single point of failure. So that is why we have built a native HHA functionality inside that VM to making sure that if one of these VMs fail, that automatically another OVM, that's how we call it, can take over the responsibilities. And again, because the drivers are stateless, it's extremely easy for us to configure it like that. We have support for most of the open stack versions out there. So we have drivers for Kilo, Liberty and Mitaka. And for the Newton piece, we today have drivers in tech preview mode and we expect these drivers to go out in the next few weeks as well. I'm sorry. So very brief, why should you run open stack on enterprise cloud platform powered by Nutanix? One, because of the flexibility. One of the top reasons in the last open stack survey, why customers are looking into open stack is because of the flexibility. Because you can run the OS of Nutanix on multiple types of hardware, as well as multiple hypervisors, we just double click on that promise as well. Continuous innovation, something that was discussed during the keynote as well, this is what we have done over the last five years within Nutanix. Every year, more or less, we release two major software versions and in each version, you will get new features, new functions that will empower your IT services to do more within your organization. Now, it is all good that we at Nutanix engineer lots of new functionalities and features, but it is nothing when customers do not use it. So we have tracked the adoption of one of our latest releases. And what you see is that more than 40% of our customers adopted to the latest and greatest software within 100 days. This is an unseen metric in a traditional data center. And why can we do that? It is because we have a technology which is called one click, one click upgrade, whereby every single component in your data center can be upgraded, either to an API call or with a single click, going from the BIOS, the hypervisor, the Nutanix OS itself, all again with a single click without interfering with your end user services. So no more need for a downtime weekend or maintenance weekend or whatsoever. Everything can be done on the fly with a single click or two on API call. Operational efficiency, again one of the promises of OpenStack. Now, what Ashish will show you in two minutes from now is how we represent our cluster into OpenStack. It is a single Nova compute object. And we within Nutanix, we will take care of the scheduling of all your instances. Not only during the creation, but also when your users are running applications on top of it, based on resource contention, we will move workloads across the cluster. You do not have to babysit that anymore with Nutanix. We also offer, like I mentioned, data center planning tools. So with a single click, you can plan for the next six or 12 months your data center. And the reason why we do that is we do not want you to have to pay for resources that you do not use today. You pay for what you need today. And whenever you need resources, you add them in your OpenStack environment. You can see some screenshots here how we make life easier from Prism. This is a graph that explains you the IO characteristics throughout your OpenStack environment. So we will indicate, okay, this is the amount of rides, reeds, that type of block size during the day, during the last weeks, and so on. And then some screenshots of our planning tools. On the left-hand side, you can see that we have 28 days left with the resources in my cloud so that you can take actions today. And action can be, look at the oversized virtual machines or instances, or maybe you just add some x86 or storage resources with a single click in the Nutanix environment. As soon as you add resources into your cluster, the Nutanix cluster, automatically, it also gets represented in OpenStack. So you don't have to do anything in OpenStack if you want to scale out or scale down your environment. Right, so let's now go to the very interesting piece and that is the demo given by Ashish. So thanks, Steven. There was very nice information about how Nutanix works and how we integrate with OpenStack. So for the demo, I'll be showing you a demo video. There are three key features which are important takeaways. I would like you guys to take away from this demo. Firstly, I have spun up an OpenStack control plane which has our Cinder, Glantz and Neutron drivers. We have a Nutanix OpenStack VM which has our compute drivers. So this together as an OpenStack control plane is talking to our cluster. Our cluster comprises of 24 nodes. And since we are a hyperconverse platform, in the demo I'll show you how we pull the resources of memory, vCPU and storage and provide it as one single hypervisor instance to OpenStack. By this, we are offloading OpenStack complexities to have scheduling policies, scheduling decisions being made at OpenStack level where we leverage Acropolis Dynamics scheduler to place the VMs. We leverage Nutanix distributed file system to access the Glantz repository or Cinder repository. And then lastly, I would like to reiterate what Stephen mentioned. So this together as a cluster has highly available cluster virtual IP. With that, you get the benefit of upgrade going on but your control plane remains untouched. Say if your VMs migrate from one hypervisor to other hypervisor to the control plane, it's seamless, the VMs are highly available and you can query the state of the VMs, the nodes, anything which is highly available for OpenStack. So let's jump into the demo. So this is our Nutanix Prism interface which is highly available. Let's just log into that. So sorry, I have blurred out some things because of some NDAs but I'll show you the important parts of it. As you can see, there are like 24 nodes in the cluster and this is the highly virtual available VIP OpenStack is talking to. Now let's go to the horizon dashboard and now we'll see how the 24 node cluster is represented in OpenStack. The aggregate of memory, the VCPU and storage resources are shown as one single compute node or one hypervisor to OpenStack. So to speed up the demo, what I've done is I've already imported the images and created the networks and in the demo, I'll be spinning off 100 VMs and show you how fast is it to provision 100 VMs on our Nutanix cluster. So I'll be using the CentOS image and a network and yeah, so this is the network configuration which has like VLAN 80 and now let's go to horizon and spin off 100 VMs. All right, so in the background, what is happening is we are, so when VM spawn call comes, we go to glance since our distributed storage fabric already has that image and all the hypervisors are aware of that image. There's no additional need for any hypervisor to download the image separately. Hence, any hypervisor which has the VM on it spins off the VMs really fast. We also acknowledge and honor Neutron's IPAM. So we get the MAC address, we get the IP address, we construct the proto of the VM, we construct the VM config, we bind the image reference and send a single API call to the cluster and that's the reason in the end we'll see how fast is it to deploy 100 VMs in Nutanix cluster. So I'm checking back and forth between the horizon UI versus Nutanix UI. Okay, so now I'll be going to the Nutanix UI to check the 100 VM, which is being spin up. As you can see, it's taking between one to two minutes for the build and schedule to happen and now let us go to the Nutanix UI and check for this particular center's VM 100. Let's check if it's booting up, there it is. And let's go to horizon back and check if the VM has booted up. So as you could see, we were like extreme, since our drivers are completely lightweight, extremely agile and fast, you could see the 100 VMs booting up within a matter of one or two minutes. So that's my demo, thanks. All right, thank you Ashish and we're just on time. So if you would like to hear more about Nutanix and how we deliver on the promises of OpenStack, just come by, we're over there. Happy to have a chat with you in depth. Thank you.