Configuring Hyper-V for High Availability

Running virtualized operating systems introduces a dependency on the underlying virtualization infrastructure. If the virtualization host is a standalone machine, it actually becomes a single point of failure for all guest systems running on top of it. So, let's get this thing clustered! But how do we do that and do so in a way that allows every single guest system to move around in that cluster independently from all the others? Well, turning one single virtual machine stored on a shared LUN into a Cluster resource has become almost embarrassingly simple with Hyper-V. But that's one - how about one hundred or one thousand? We might want to avoid using drive letters in that case - unless you want to limit yourself to 22 virtual machines on a 16 node cluster. How then do we store the guest systems data and configuration files? Actually there are multiple ways of how to accomplish this, and we'll look into the technical details, and the advantages and disadvantages of those, and try them out live on stage.