Chapter 13: Scaling out KVM with AWS

Virtualization is a hard problem if you take a look at it up close – it is complicated to emulate complete computers in order to be able to run operating systems on them. For obvious reasons, getting those virtual machines into a cloud is even harder. After that, things really start to get messy. Conceptually, creating clusters of machines that can run on demand is even more complicated, based on the sheer number of machines that must run at the same time. Also, there's a need to create not only emulated computers but also all the networking and infrastructure to support larger deployments. Creating a global cloud – one that not only runs millions of machines but is also almost omnipresent in even the most remote parts of the globe – is a task that few companies have ever tried, and only a couple have succeeded. This chapter will cover those big cloud providers in general, and then Amazon, as the biggest of them all. Our...