
Pity the Performance
It might seem like a nice idea to have everything as a virtualized stack but how many context switches is an I/O going to need as it passes through multiple virtual machines on the same CPU to get from "server" to "storage"? There's only so much CPU power you can use to compensate for this type of massively inefficient design, and in a world where SSD-level performance is expected it just isn't going to be able to keep up.