Date: Tue, 20 Apr 2010 11:46:14 -0700
From: David Kerr <dmk@mr-paradox.net>
To: Kevin Grittner <Kevin.Grittner@wicourts.gov>
Cc: Robert Haas <robertmhaas@gmail.com>, pgsql-performance@postgresql.org
Subject: Re: Very high effective_cache_size == worse performance?
Message-ID: <20100420184614.GD53489@mr-paradox.net>
References: <20100420173936.GA50886@mr-paradox.net>
	<h2w603c8f071004201044tbc8ce8fenb846b7643a2b9376@mail.gmail.com>
	<20100420180351.GA51940@mr-paradox.net>
	<4BCDA94E0200002500030BF8@gw.wicourts.gov>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <4BCDA94E0200002500030BF8@gw.wicourts.gov>
User-Agent: Mutt/1.4.2.3i

On Tue, Apr 20, 2010 at 01:17:02PM -0500, Kevin Grittner wrote:
- David Kerr <dmk@mr-paradox.net> wrote:
-  
- > Incidentally the code is written to work like this :
- > 
- > while (read X lines in file){
- > Process those lines.
- > write lines to DB.
- > }
-  
- Unless you're selecting from multiple database tables in one query,
- effective_cache_size shouldn't make any difference.  There's
- probably some other reason for the difference.
-  
- A couple wild shots in the dark:
-  
- Any chance the source files were cached the second time, but not the
- first?
-  
- Do you have a large checkpoint_segments setting, and did the second
- run without a new initdb?
-  
- -Kevin

no i don't think the files would be cached the 2nd time. I ran it multiple times
and got the same performance each time. It wasn't until i changed the parameter
that performance got better.

I've got checkpoint_segments = 300

Dave