...building hybrid clouds that can support any device from anywhere
“How do you turn a stack of hardware into a fully functioning “Infrastructure-on-Demand” service as quickly and cheaply as possible?” That was the question confronting our team as we brainstormed our next project. Before I dive into that loaded question, let’s provide some quick background on part of what our team does.
Glad you asked! For several years now, we have been responsible for an environment called SCDemo (Rob mentioned this in an earlier post with his Deployment Track). Our goal when creating that environment was both broad and challenging: Enable Microsoft’s technical field employees all over the world to use our wide array of enterprise server software, without requiring them to invest in expensive hardware. Our solution? Host it for them, while giving them full control to replicate the experience of managing their own miniature datacenters.
Our SCDemo program was such a great success that we decided to see if we could extend its reach to an even wider group of employees.
As we developed this program we kept something simple in mind: Dedicated hardware is expensive, and isn’t necessary for everyone who wants to demonstrate the power of Windows Server, System Center, and our cloud solutions, therefore, our next step was to practice what we preach and provide a real self-service infrastructure experience based on Windows Server 2012 and System Center 2012 SP1.
The next few posts will deal with the basics of infrastructure: Hosts, networking, storage. Following this analysis, I’ll explore how we built the service on top of it. While building SCDemo one of our primary goals was to do this as simply as possible, with as little (or no) custom development work, while using off-the-shelf Microsoft components. This approach has created a wealth of information and techniques that are valuable to everyone using our products, and I’m confident it can help in the design of your own infrastructure services.