Performance increase for Koji builders

[Miroslav Suchy]

Dne 2.6.2015 v 00:16 Kevin Fenzi napsal(a):
> The thing I find odd about this is that the linux kernel caching
> doesn't seem to be a win. Shouldn't it cache buildroot and such in
> memory anyhow? Or why is tmpfs performing so much better?

When building mock.src.rpm it allocated about 1.5GB in tmpfs. Never done
any disk IO. When you rely on kernel caching it will still sync() every
few seconds (is it still 30 secs?). So it will write those 1.5GB to
disk. However I even seen a lot of reading in dstat output.

Mirek