>This an 8 core opteron 2.1GHz machine with 32G ram and a 12 disk RAID-10 f=
or pgdata >and a two disk RAID-1 for pg_xlog.


OK, based on that information, I can only suggest the following tweaks. Try=
 them

1 at a time to see what difference they make. Note that shared_buffers & ma=
x_files_per_process require a restart. The others can be handled from psql =
with =


SELECT pg_reload_conf();



shared_buffers=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0 -->=
 increase to 48mb

work_mem=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=
=A0=A0=A0 --> increase to 8mb

maintenance_work_mem=A0=A0 --> increase to 32mb

max_files_per_process=A0=A0=A0=A0=A0 --> increase to 4000



Also, check the postgresql log to see if there are any warnings about
the need to increase other parameters such as max_fsm_pages or
checkpoint_segments.



Good luck,


Melvin Davidson =





      =

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.slony.info/pipermail/slony1-general/attachments/20091108/=
50ada08c/attachment-0001.htm