 Intel IT, we've been running a very large-scale grid computing environment. In fact, that's my background, where we run a lot of silicon design over a Linux grid across the globe. About four years ago, we made a decision to run an enterprise private cloud for traditional IT. We're pretty successful. We've had it running since 2010, but we made a call a couple years ago based on our large background with open source that moving forward, we're going to do green field environments focused on open source solutions. When we focused on private cloud initially, it was basically with a concept that since we have such a large scale inside of our environment, we felt that based off of cost, security and performance, it made a lot of sense for us to continue down this path of private cloud. However, as the world has changed and clouds become more mature, we've come to recognize that hybrid cloud is really the way to go. This is because we have 68 data centers, but I don't have data centers everywhere across the globe. I really see cloud as part of the supply chain. As we push forward with this hybrid offering, I want to be able to use these open APIs. If I need a data center capacity in South America, I can use one of my partners in the public cloud space. Preferably, they're using open APIs, so my developers can write once. That's really where we're going is pushing in the hybrid state and allowing our developers to write once and utilize public or private and where we help control based off cost, performance and location where they're getting their compute and storage capacity from. About two years ago, we basically looked at all the solutions in an environment that was happening and we made a choice and basically a bet on OpenStack. Ever since then, we've been pretty successful running this in a production mode since mid-2012 and we haven't looked back, so far working great. Fundamentally, the reason that we went after OpenStack is we're trying to turn our entire data center basically into software, so all the infrastructure needs to be exposed as APIs. With OpenStack, we've been running with Essex focusing on compute predominantly, but as we move forward into the next couple months, we'll be turning on object storage and block storage to basically have a completely robust data center infrastructure exposed as APIs. With cloud, it's a great opportunity to really create an ecosystem that allows soft developers to basically run in a federated fashion across the globe. I see cloud as part of my supply chain whether it's private or public and by turning on OpenAPIs, we're really able to basically have a game-changer where the software developers can write once to many types of clouds. So the reason that we really focus on the OpenAPIs is how do we drive to a federated environment that's interoperable, meaning that they can work across any cloud across the world, we're a global company, and then three, how do we do it in open fashion? So these three elements together are very important to us, which is why we're taking such a specific step into the OpenStack space. So eventually, we'd like to just stop saying hybrid cloud, there's no public, there's no private, it's just this very large federated interoperable open cloud. You can really almost see it as utility, finally, where you're really just getting electricity, but where you're getting is compute and storage globally across the world. Whether we're doing business in China, South America and the United States, we need to operate everywhere. So having this all federated so we can connect together with similar identities, similar security, interoperable where we can basically write once or control it through our infrastructure teams with one set of controls, or open, which is really how to get there by enabling the industry to move up the stack from the core of Linux to now have open APIs in the cloud space and really allows a level of innovation to start happening higher and higher up the stack and allows us to run it as a global industry.