| From: | Jim Nasby <decibel(at)decibel(dot)org> |
|---|---|
| To: | Greg Smith <gsmith(at)gregsmith(dot)com> |
| Cc: | pgsql-hackers(at)postgresql(dot)org |
| Subject: | Re: Log levels for checkpoint/bgwriter monitoring |
| Date: | 2007-03-09 21:34:49 |
| Message-ID: | E440FC7A-8148-41FB-BF56-BB651603D4C1@decibel.org |
| Views: | Whole Thread | Raw Message | Download mbox | Resend email |
| Thread: | |
| Lists: | pgsql-hackers |
On Mar 8, 2007, at 11:51 PM, Greg Smith wrote:
> almost everything that's dirty is also pinned during pgbench, and
> the LRU is lucky to find anything it can write as a result
I'm wondering if pg_bench is a good test of this stuff. ISTM it's
unrealistically write-heavy, which is going to tend to not only put a
lot of dirty buffers into the pool, but also keep them pinned enough
that you can't write them.
Perhaps you should either modify pg_bench to do a lot more selects
out of the various tables
--
Jim Nasby jim(at)nasby(dot)net
EnterpriseDB http://enterprisedb.com 512.569.9461 (cell)
| From | Date | Subject | |
|---|---|---|---|
| Next Message | Jim Nasby | 2007-03-09 21:38:10 | Re: Log levels for checkpoint/bgwriter monitoring |
| Previous Message | Jim Nasby | 2007-03-09 21:28:48 | Re: Log levels for checkpoint/bgwriter monitoring |