Re: Help with tuning this query (with explain analyze finally) - Mailing list pgsql-performance

From Josh Berkus
Subject Re: Help with tuning this query (with explain analyze finally)
Date
Msg-id 200503030935.14537.josh@agliodbs.com
Whole thread Raw
In response to Re: Help with tuning this query (with explain analyze finally)  ("Ken Egervari" <ken@upfactor.com>)
List pgsql-performance
Ken,

Well, I'm a bit stumped on troubleshooting the actual query since Windows'
poor time resolution makes it impossible to trust the actual execution times.
Obviously this is something we need to look into for the Win32 port for
8.1 ..

> shared_buffers = 1000

This may be slowing up that merge join.  Try resetting it to 6000.    I'm not
sure what system settings you might have to do on Windows to get it to
support higher shared buffers; see the docs.

> #work_mem = 1024

Up this to 4096 for testing purposes; your production value will vary
depending on several factors; see link below.

> #effective_cache_size = 1000

Increase this to the actual amount of RAM you have available, about 750MB (you
do the math)

> #random_page_cost = 4

Leave this for now.

See www.powerpostgresql.com/PerfList for more information.

--
Josh Berkus
Aglio Database Solutions
San Francisco

pgsql-performance by date:

Previous
From: Jeremiah Jahn
Date:
Subject: Re: name search query speed
Next
From: Josh Berkus
Date:
Subject: Re: name search query speed