Linux – Feeding the kernels entropy source from other machines and/or increasing its maximum size

Like you said. If the journal is bigger you have more possibilities with that filesystem but I don't really think that's the only way you're getting the better performance. You don't need a partition to try the new ext4 on a filesystem. You can create an image file with the dd command like so : Create a 1 GiB file containing only zeros (bs=blocksize, count=number of blocks):

dd if=/dev/zero of=file1G.tmp bs=1M count=1024

then you can create a ext4 filesystem in that file:

mkfs.ext4 /path/to/file1G.tmp

I would say that unless you are concerned about a determined and sophisticated attacker with a great deal of resources compromising your site by breaking the pseudo-random algorithm used in the Linux kernel, urandom is probably okay.

You'll find that anything encryption related, including cryptographic hashing (salts are random, even if the hashing function doesn't involve randomness itself), will deplete your supply of entropy. Also, a virtual machine will tend to have less entropy, as entropy is acquired from a variety of things that VMs just don't have, or can't securely get entropy from (e.g. physical mouse, keyboard, or clock).

I'm also not sure why you consider 100-200 a small amount. I'd consider 0-10 to be small, but keep in mind the maximum size of the pool (/proc/sys/kernel/random/poolsize) may be as little as 4096. Unless you're at 0 a fraction of the time, you're probably not using it up too fast (you might already be using urandom without knowing it). As best I know, reads from /dev/random won't block until the entropy pool is 0.