Paulius Pazera wrote:

> I thought I share my experience. Brief conclusion - avoid ext2/ext3 file
> systems for big databases, for example XFS is significantly better

It is known that XFS performs much better with big files. It is great
for backup and restore as a least one file (backup file) is read
sequentially. However, I'm not sure backup and restore are the right way
to "benchmark" a filesystem for database storage. Try performing a heavy
query that would read database at many places (not sequentially as
restore does) and see what happens. I'm not sure the difference would be
that big.

I would also like to see parallel test for ReiserFS which is the default
on many distributions (and maybe JFS). Any chance you do one of those?

--
Milan Babuskov
http://www.flamerobin.org