Message-ID: <43614D9F.9010600@pendragon.org>
Date: Thu, 27 Oct 2005 15:58:55 -0600
From: Michael Best <mbest@pendragon.org>
User-Agent: Mozilla/5.0 (X11; U; Linux i686; en-US;
	rv:1.7.12) Gecko/20050922 Fedora/1.7.12-1.3.1
MIME-Version: 1.0
To: pgsql-performance@postgresql.org
Subject: Re: how postgresql request the computer resources
References: <BAY23-F28F5DD23A4A73C7BFE3CD1CE680@phx.gbl>
	<4360F8C1.4040202@archonet.com>
In-Reply-To: <4360F8C1.4040202@archonet.com>
Content-Type: text/plain; charset=us-ascii; format=flowed
Content-Transfer-Encoding: 7bit

Richard Huxton wrote:
>> WITH ANY OF THIS QUERIES MSSQL TAKES NOT MUCH OF 7 SECONDS....
> 
> 
> In which case they make a bad choice for showing PostgreSQL is faster 
> than MSSQL. Is this the only query you have, or are others giving you 
> problems too?
> 
> I think count(*) is about the weakest point in PG, but I don't think 
> there'll be a general solution available soon. As I'm sure someone has 
> mentioned, whatever else, PG needs to check the row for its visibility 
> information.
> 
>  From the start of your email, you seem to suspect your configuration 
> needs some work. Once you are happy that your settings in general are 
> good, you can override some by issuing set statements before your query. 
> For example:
>     SET work_mem = 10000;
> might well improve example #2 where you had a hash.
> 
> -- 
>   Richard Huxton
>   Archonet Ltd

Someone had suggested keeping a vector table with +1 and -1 for row 
insertion and deletion and then running a cron to sum the vectors and 
update a table so that you could select from that table to get the row 
count.  Perhaps some sort of SUM() on a column function.

Since this seems like a reasonable approach (or perhaps there may be yet 
another better mechanism), cannot someone add this sort of functionality 
to Postgresql to do behind the scenes?

-Mike