Re: Tuning/performance issue....

From: Shridhar Daithankar <shridhar_daithankar(at)persistent(dot)co(dot)in>
To: David Griffiths <dgriffiths(at)boats(dot)com>
Cc: pgsql-performance(at)postgresql(dot)org
Subject: Re: Tuning/performance issue....
Date: 2003-10-01 07:05:30
Message-ID: 3F7A7CBA.4060507@persistent.co.in
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-performance

David Griffiths wrote:

> And finally,
>
> Here's the contents of the postgresql.conf file (I've been playing with
> these setting the last couple of days, and using the guide @
> http://www.varlena.com/varlena/GeneralBits/Tidbits/annotated_conf_e.html to
> make sure I didn't have it mis-tuned):
>
> tcpip_socket = true
> max_connections = 500 # We will need quite a few connections;
> currently only one connection to database, however
> port = 5432
> shared_buffers = 5000 # I've tried 5000 to 80,000 with no
> apparent difference
> wal_buffers = 16
> sort_mem = 256 # decreased this due to the large # of
> connectiosn
> effective_cache_size = 50000 # read that this can improve performance;
> hasn't done anything.

Reading this whole thread, I think most of the improvement you would get would
be from rethinking your schema from PG point of view and examine each query.

After you changed your last query as Tom suggested for explicit join, how much
improvement did it make? I noticed that you put
'commercial_entity.commercial_entity_id=225528' as a second codition. Does it
make any difference to put it ahead in where clause list?

HTH

Shridhar

In response to

Browse pgsql-performance by date

  From Date Subject
Next Message Andrew Sullivan 2003-10-01 10:33:38 Re: inferior SCSI performance
Previous Message David Griffiths 2003-10-01 05:48:54 Re: Tuning/performance issue...