Re: Postgres 7.3.1 poor insert/update/search performance - Mailing list pgsql-performance

From Stephan Szabo
Subject Re: Postgres 7.3.1 poor insert/update/search performance
Date
Msg-id 20030121134242.Q84028-100000@megazone23.bigpanda.com
Whole thread Raw
In response to Postgres 7.3.1 poor insert/update/search performance esp WRT Oracle  ("Seth Robertson" <pgsql-performance@sysd.com>)
Responses Re: Postgres 7.3.1 poor insert/update/search performance
List pgsql-performance
On Tue, 21 Jan 2003, Seth Robertson wrote:

> The postgresql.conf file is default (my sysadmin nuked all of my
> changes when he upgraded to 7.3.1--grr) and there are some shared
> memory configs: kernel.sem = 250 32000 100 128, kernel.shmmax =
> 2147483648, kernel.shmmni = 100, kernel.shmmax = 134217728  The
> WAL is not seperated (but see below).

You almost certainly want to raise shared_buffers from the default (64?)
to say 1k-10k.  I'm not sure how much that'll help but it should help
some.

> A "vacuum analyze" is performed between the write phase and the read
> phase.  However, for your analysis pleasure, here are the results
> of a full verbose analyze and some explain results (both before and after).

BTW: what does explain analyze (rather than plain explain) show?


pgsql-performance by date:

Previous
From: "Seth Robertson"
Date:
Subject: Postgres 7.3.1 poor insert/update/search performance esp WRT Oracle
Next
From: Seth Robertson
Date:
Subject: Re: Postgres 7.3.1 poor insert/update/search performance