Message-ID: <4604062D.3000905@wildenhain.de>
Date: Fri, 23 Mar 2007 17:54:05 +0100
From: Tino Wildenhain <tino@wildenhain.de>
User-Agent: Thunderbird 1.5.0.10 (Windows/20070221)
MIME-Version: 1.0
To: pgsql-performance@postgresql.org
Subject: Re: Performance of count(*)
References: <Pine.LNX.4.62.0703221149210.13747@wr-linux02>
	<200703221208.19358.albert@sedifa.com>
	<Pine.LNX.4.64.0703221323590.1503@ismoli.solenovo.jns>
	<20070322141807.GE11402@mathom.us>
	<46029DE3.9030505@modgraph-usa.com>
	<4602A15B.7000908@wildenhain.de>
	<4602BB19.9050306@modgraph-usa.com>
	<4602BC84.8030101@wildenhain.de>
	<20070322173930.GI11402@mathom.us> <4603C17E.709@wildenhain.de>
	<20070323130910.GL11402@mathom.us>
In-Reply-To: <20070323130910.GL11402@mathom.us>
Content-Type: text/plain; charset=ISO-8859-1; format=flowed
Content-Transfer-Encoding: 7bit

Michael Stone schrieb:
> On Fri, Mar 23, 2007 at 01:01:02PM +0100, Tino Wildenhain wrote:
>> This discussion is a bit theoretical until we see the actual problem
>> and the proposed solution here.
> 
> It's good to see you back off a bit from your previous stance of 
> assuming that someone doesn't know what they're doing and that their 
> solution is absolutely wrong without actually knowing anything about 
> what they are trying to do.

Well I'm sure its very likely wrong :-) At least the core part of
it with the statement of "keeping index data outside postgres".

What I meant with my comment about the theoreticalness: we cannot
make educated suggestions about alternative solutions to the problem
until we know the problem and maybe the current solution in detail.

Regards
Tino