| From: | Jeff Davis <pgsql(at)j-davis(dot)com> |
|---|---|
| To: | Chris Hoover <revoohc(at)gmail(dot)com> |
| Cc: | PGSQL Performance <pgsql-performance(at)postgresql(dot)org> |
| Subject: | Re: Is this true? |
| Date: | 2007-06-06 21:37:04 |
| Message-ID: | 1181165824.7660.157.camel@dogma.v10.wvs |
| Views: | Raw Message | Whole Thread | Download mbox | Resend email |
| Thread: | |
| Lists: | pgsql-performance |
On Wed, 2007-06-06 at 16:58 -0400, Chris Hoover wrote:
> Question,
>
> Does (pg_stat_get_db_blocks_fetched(oid)-pg_stat_get_db_blocks_hit
> (oid)*8) = number of KB read from disk for the listed database since
> the last server startup?
That will give you the number of blocks requested from the OS. The OS
does it's own caching, and so many of those reads might come from the OS
buffer cache, and not the disk itself.
Also, if you're concerned with the number since the last server restart,
make sure you have stats_reset_on_server_start set appropriately.
Regards,
Jeff Davis
| From | Date | Subject | |
|---|---|---|---|
| Next Message | Dan Harris | 2007-06-06 22:04:44 | reclaiming disk space after major updates |
| Previous Message | Scott Marlowe | 2007-06-06 21:27:47 | Re: VERY slow queries at random |