Otto, Separate the pg_xlog directory onto its own filesystem and retry your tests.
Bob Lunney ________________________________ From: Havasvölgyi Ottó <havasvolgyi.o...@gmail.com> To: Marti Raudsepp <ma...@juffo.org> Cc: Aidan Van Dyk <ai...@highrise.ca>; pgsql-performance@postgresql.org Sent: Thursday, December 8, 2011 9:48 AM Subject: Re: [PERFORM] Response time increases over time I have moved the data directory (xlog, base, global, and everything) to an ext4 file system. The result hasn't changed unfortuately. With the same load test the average response time: 80ms; from 40ms to 120 ms everything occurs. This ext4 has default settings in fstab. Have you got any other idea what is going on here? Thanks, Otto 2011/12/8 Marti Raudsepp <ma...@juffo.org> On Thu, Dec 8, 2011 at 06:37, Aidan Van Dyk <ai...@highrise.ca> wrote: >> Let me guess, debian squeeze, with data and xlog on both on a single >> ext3 filesystem, and the fsync done by your commit (xlog) is flushing >> all the dirty data of the entire filesystem (including PG data writes) >> out before it can return... > >This is fixed with the data=writeback mount option, right? >(If it's the root file system, you need to add >rootfsflags=data=writeback to your kernel boot flags) > >While this setting is safe and recommended for PostgreSQL and other >transactional databases, it can cause garbage to appear in recently >written files after a crash/power loss -- for applications that don't >correctly fsync data to disk. > >Regards, >Marti >