June 20, 2013

Comment: Linux schedulers in tpcc like benchmark

… short recap with some numbers (40 simultaneous seq reads): XFS+noop: 140MB/s EXT3+noop: 133MB/s EXT4+noop: 111MB/s (exploded once with corrupt data when changing scheduler) EXT2+noop: 97MB/s XFS+cfq: 30MB/s Some comments on this setup: – XFS beats EXT*, with EXT4 performing badly and also being…

Comment: Performance problem with Innodb and DROP TABLE

Robert, I wonder if you are using ext3? The performance of dropping large tables is mostly bound by IO when using ext3. XFS can drop large files much more quickly.

Post: Working with many files and file system fragmentation

… the file sequentially and reading it back. The performance difference was huge – we could read single file … is a massive difference. The test was done on EXT3 and it looks like it does not do … other filesystems deal with this problem, for example XFS with delayed allocation may be doing better job….

Post: Binary log file size matters (sometimes)

…pmp or oprofile on that system due to potential performance impact. However, somehow I have noticed that pt-…slow at deleting files (both customers systems ran on ext3, which can be very slow deleting large files), … long to remove. Note that on ext4 and xfs you should not have this problem as they …