Re: Volunteer to build a configuration tool

Dave Page <dpage@postgresql.org>

From: Dave Page <dpage@postgresql.org>
To: Tom Lane <tgl@sss.pgh.pa.us>
Cc: Greg Smith <gsmith@gregsmith.com>, pgsql-performance@postgresql.org
Date: 2007-06-21T15:09:07Z
Lists: pgsql-performance
Tom Lane wrote:
> There's another reason for not setting shared_buffers huge, beyond the
> good ones Greg listed: the kernel may or may not consider a large
> shared-memory segment as potentially swappable.  

Another is that on Windows, shared memory access is more expensive and 
various people have noted that the smallest value for shared_buffers you 
can get away with can yield better performance as it leaves more free 
for the kernel to use, more efficiently.

Regards, Dave.