Ok, it is now time to test some real-world applications. The first on the list is the mysql database system, benchmarked using Sysbench mysql test module.
Lets see how much time each filesystem requires to create a 100K rows test database:
All filesystems perform similarly here. What happen if we begin to hammer the database with thousand of read (select) requests?
Again, all filesystems perform more or less the same. What about the much more intensive, read-write transactional test?
BTRFS is quite slow here.
From Sysbench tests, it seems that, apart btrfs, all filesystems will be valid choices for mysql workloads. Will the integrated mysql-bench benchmark confirm this statement?
Yes: while XFS seems a little slower then ext3/4, the difference is quite narrow. On the other side, BTRFS is considerably slower then the others.
So, in the end, the real lesson here is that btrfs should be avoided in write-intensive mysql workloads.