| From: | Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us> |
|---|---|
| To: | Ron <rjpeace(at)earthlink(dot)net> |
| Cc: | Matthew Sackman <matthew(at)lshift(dot)net>, pgsql-performance(at)postgresql(dot)org |
| Subject: | Re: Massive performance issues |
| Date: | 2005-09-01 20:25:29 |
| Message-ID: | 6092.1125606329@sss.pgh.pa.us |
| Views: | Raw Message | Whole Thread | Download mbox | Resend email |
| Thread: | |
| Lists: | pgsql-performance |
Ron <rjpeace(at)earthlink(dot)net> writes:
> ... Your target is to have each row take <= 512B.
Ron, are you assuming that the varchar fields are blank-padded or
something? I think it's highly unlikely that he's got more than a
couple hundred bytes per row right now --- at least if the data is
what it sounds like.
The upthread comment about strcoll() set off some alarm bells in my head.
If the database wasn't initdb'd in C locale already, try making it so.
Also, use a single-byte encoding if you can (LatinX is fine, Unicode not).
> Upgrade pg to 8.0.3 and make sure you have enough RAM for your real
> day to day load.
Newer PG definitely better. Some attention to the configuration
parameters might also be called for. I fear though that these things
are probably just chipping at the margins ...
regards, tom lane
| From | Date | Subject | |
|---|---|---|---|
| Next Message | Sebastian Hennebrueder | 2005-09-01 20:39:13 | Re: Massive performance issues |
| Previous Message | Luke Lonergan | 2005-09-01 20:24:30 | Re: Poor performance on HP Package Cluster |