| From: | Erik Jones <ejones(at)engineyard(dot)com> |
|---|---|
| To: | wstrzalka <wstrzalka(at)gmail(dot)com> |
| Cc: | pgsql-general(at)postgresql(dot)org |
| Subject: | Re: Pet Peeves? |
| Date: | 2009-02-05 23:05:24 |
| Message-ID: | F42784A8-2A68-4E11-8501-960C23909A58@engineyard.com |
| Views: | Whole Thread | Raw Message | Download mbox | Resend email |
| Thread: | |
| Lists: | pgsql-general |
On Feb 2, 2009, at 12:46 PM, wstrzalka wrote:
> On Feb 2, 8:23 pm, br(dot)(dot)(dot)(at)momjian(dot)us (Bruce Momjian) wrote:
>> wstrzalka wrote:
>>> * stat collector is really greedy by definition even when system is
>>> idle, when you have really really many relations
>>
>> I think this will be fixed in 8.4.
>>
>
> That would by great news for "mine" cluster.
One workaround I came up with a while back for that is to edit the
stat file name to be in a separate directory under global (like
<datadir>/global/pg_stats/pgstat.stat) and mount a ramfs there. Of
course, a custom compile isn't always an option but it removed a *ton*
of IO on that db (had thousands upon thousands of tables). Also, if
you do that you need to be sure to copy pgstat.stat to a permanent
place periodically unless you want to risk losing all of your stats.
Erik Jones, Database Administrator
Engine Yard
Support, Scalability, Reliability
866.518.9273 x 260
Location: US/Pacific
IRC: mage2k
| From | Date | Subject | |
|---|---|---|---|
| Next Message | Gregory Stark | 2009-02-05 23:10:35 | Re: Bringing other columns along with a GROUP BY clause |
| Previous Message | Rob Richardson | 2009-02-05 21:31:40 | Re: Bringing other columns along with a GROUP BY clause |