Our disk service times and % utilization (according to sar -d) while running pg_dump are low. For example: 01:23:08 AM DEV tps rd_sec/s wr_sec/s avgrq-sz avgqu-sz await svctm %util 01:23:09 AM sda 1473.00 0.00 98128.00 66.62 0.41 0.28 0.03 3.70 01:23:09 AM sda1 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 01:23:09 AM sda2 1473.00 0.00 98128.00 66.62 0.41 0.28 0.03 3.70 01:23:09 AM sdb 1.00 0.00 16.00 16.00 0.00 3.00 3.00 0.30 01:23:09 AM sdb1 1.00 0.00 16.00 16.00 0.00 3.00 3.00 0.30 01:23:09 AM sdc 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 01:23:09 AM sdd 678.00 161584.00 0.00 238.32 0.34 0.49 0.45 30.20 01:23:09 AM sdd1 678.00 161584.00 0.00 238.32 0.34 0.49 0.45 30.20 01:23:09 AM nodev 12266.00 0.00 98128.00 8.00 4.30 0.35 0.00 3.70 01:23:09 AM nodev 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 On Fri, Feb 25, 2011 at 3:50 PM, Benjamin Smith <lists@xxxxxxxxxxxxxxxxxx> wrote: > I'd also add: run pg_tune on your server. Made a *dramatic* difference for > us. Well, that's interesting. I am going to try that, thanks!! Aleksey. -- Sent via pgsql-general mailing list (pgsql-general@xxxxxxxxxxxxxx) To make changes to your subscription: http://www.postgresql.org/mailpref/pgsql-general
