Robert Haas <robertmhaas@gmail.com> wrote:
> Then again, is this a regular pgbench test or is this SELECT-only?
SELECT-only
> Can you by any chance check top or vmstat during the 32-client
> test and see what percentage you have of user time/system
> time/idle time?
You didn't say whether you wanted master or flexlock, but it turned
out that any difference was way too far into the noise to show.
They both looked like this:
procs --------------memory------------- ---swap-- -----io----r b swpd free buff cache si so bi
bo----system---------cpu------ in cs us sy id wa st
38 0 352 1157400 207177020 52360472 0 0 0 16 13345 1190230 40 7 53 0 0
37 0 352 1157480 207177020 52360472 0 0 0 0 12953 1263310 40 8 52 0 0
36 0 352 1157484 207177020 52360472 0 0 0 0 13411 1233365 38 7 54 0 0
37 0 352 1157476 207177020 52360472 0 0 0 0 12780 1193575 41 7 51 0 0
Keep in mind that while there are really 32 cores, the cpu
percentages seem to be based on the "threads" from hyperthreading.
Top showed pgbench (running on the same machine) as eating a pretty
steady 5.2 of the cores, leaving 26.8 cores to actually drive the 32
postgres processes.
> What OS are you running?
Linux new-CIR 2.6.32.43-0.4-default #1 SMP 2011-07-14 14:47:44 +0200
x86_64 x86_64 x86_64 GNU/Linux
SUSE Linux Enterprise Server 11 (x86_64)
VERSION = 11
PATCHLEVEL = 1
-Kevin