| From: | Richard Huxton <dev(at)archonet(dot)com> |
|---|---|
| To: | "Louis-Marie Croisez" <louis-marie(dot)croisez(at)etca(dot)alcatel(dot)be> |
| Cc: | <pgsql-general(at)postgresql(dot)org> |
| Subject: | Re: access time performance problem |
| Date: | 2002-10-10 14:29:10 |
| Message-ID: | 200210101529.10533.dev@archonet.com |
| Views: | Raw Message | Whole Thread | Download mbox | Resend email |
| Thread: | |
| Lists: | pgsql-general |
On Thursday 10 Oct 2002 12:40 pm, Louis-Marie Croisez wrote:
> Hi Richard,
> I have test it right now mounting the ext3 partition with ext2.
> The result is identical.
Hmm - that's ruled the journalling out then. Logically, it can only be two
things:
1. PostgreSQL flushing buffers or similar.
2. Another process flushing buffers or similar.
Can you try it updating batches of 200 records rather than 1000 - if the time
between delays increases that would indicate it's PG doing something.
Otherwise it could be related to logging, another process or the OS.
Try tracing activity with vmstat/iostat - that should show you what's
happening in a general sense. Then we can see if we can pin it down in
detail.
--
Richard Huxton
| From | Date | Subject | |
|---|---|---|---|
| Next Message | Felipe Schnack | 2002-10-10 14:36:42 | encryption |
| Previous Message | Joe Conway | 2002-10-10 14:28:00 | Re: bytea type |