This is so not my area of expertise. However, I wonder, wouldn't more small servers imply more hardware failures? Exchanging one HD in a big server might be less work than exchanging 10 HDs in small servers. Of course the HD in the big server might take down more "applications", but I think that is taken care of in other ways (hot swappable HDs, redundance?).
Overall I wonder if maybe a new OS for the cloud is called for? It seems inefficient to have seperate VMs running full OS for every tiny application. Maybe in the future not only storage will be a service (like S3), but also CPUs and RAM could be plugged together at will? Like there wouldn't be lots of small or big server instances, there would be farms of CPUs, farms of RAM, farms of storage, that could be combined at will. Maybe networking would be too much of a bottleneck, though :-/
The number of disks you have should be determined by your workload, not the number of servers. Think of it more like 12 disks in a big server vs. one disk in a small server -- the total number of disks (and disk failures) is the same.
Overall I wonder if maybe a new OS for the cloud is called for? It seems inefficient to have seperate VMs running full OS for every tiny application. Maybe in the future not only storage will be a service (like S3), but also CPUs and RAM could be plugged together at will? Like there wouldn't be lots of small or big server instances, there would be farms of CPUs, farms of RAM, farms of storage, that could be combined at will. Maybe networking would be too much of a bottleneck, though :-/