Date: Fri, 17 Jun 2005 00:46:43 -0400
From: Alvaro Herrera <alvherre@surnet.cl>
To: Todd Landfried <tlandfried@viatornetworks.com>
Cc: pgsql-performance@postgresql.org
Subject: Re: Needed: Simplified guide to optimal memory
Message-ID: <20050617044643.GA19751@surnet.cl>
References: <Pine.LNX.4.44.0506151123510.1139-100000@zigo.dhs.org>
	<2608.1118845105@sss.pgh.pa.us>
	<9A491A53-A5F5-4061-B1EA-8E4216299407@viatornetworks.com>
	<1118941260.698.5.camel@archimedes.mirlogic.com>
	<97CA7057-5DEE-4AA5-A4F3-8DEC1CAE720F@viatornetworks.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <97CA7057-5DEE-4AA5-A4F3-8DEC1CAE720F@viatornetworks.com>
User-Agent: Mutt/1.5.9i

On Thu, Jun 16, 2005 at 07:15:08PM -0700, Todd Landfried wrote:
> Thanks for the link. I'll look into those.
> 
> I'm going only on what my engineers are telling me, but they say  
> upgrading breaks a lot of source code with some SQL commands that are  
> a pain to hunt down and kill. Not sure if that's true, but that's  
> what I'm told.

This is true.  Migrating to a newer version is not a one-day thing.  But
increasing shared_buffers is trivially done, would get you lots of
benefit, and it's very unlikely to break anything.  (Migrating one
version can be painful already -- migrating three versions on one shot
might be a nightmare.  OTOH it's much better to pay the cost of
migration once rather than three times ...)

-- 
Alvaro Herrera (<alvherre[a]surnet.cl>)
"The Postgresql hackers have what I call a "NASA space shot" mentality.
 Quite refreshing in a world of "weekend drag racer" developers."
(Scott Marlowe)