Date: Sun, 20 May 2007 20:10:17 +0200
To: "Tom Lane" <tgl@sss.pgh.pa.us>
Subject: Re: Postgres Benchmark Results
From: PFC <lists@peufeu.com>
Cc: "Arjen van der Meijden" <acmmailing@tweakers.net>,
	pgsql-performance@postgresql.org
Content-Type: text/plain; format=flowed; delsp=yes; charset=utf-8
MIME-Version: 1.0
References: <op.tsmrv71ecigqcu@apollo13> <46507A40.505@tweakers.net>
	<op.tsmxycgdcigqcu@apollo13> <2438.1179681998@sss.pgh.pa.us>
Content-Transfer-Encoding: 7bit
Message-ID: <op.tsm0rfs4cigqcu@apollo13>
In-Reply-To: <2438.1179681998@sss.pgh.pa.us>
User-Agent: Opera Mail/9.10 (Linux)

On Sun, 20 May 2007 19:26:38 +0200, Tom Lane <tgl@sss.pgh.pa.us> wrote:

> PFC <lists@peufeu.com> writes:
>> 	The little curve that dives into the ground is MySQL with InnoDB.
>> 	The Energizer bunny that keeps going is Postgres.
>
> Just for comparison's sake it would be interesting to see a curve for
> mysql/myisam.  Mysql's claim to speed is mostly based on measurements
> taken with myisam tables, but I think that doesn't hold up very well
> under concurrent load.
>
> 			regards, tom lane


	I'm doing that now. Here is what I wrote in the report :

	Using prepared statements (important), Postgres beats MyISAM on "simple  
selects" as they say, as well as complex selects, even with 1 thread.
	
	MyISAM caused massive data corruption : posts and topics disappear,  
storage engine errors pop off, random thrashed rows appear in the forums  
table, therefore screwing up everything, etc. In short : it doesn't work.  
But, since noone in their right mind would use MyISAM for critical data, I  
include this result anyway, as a curiosity.

	I had to write a repair SQL script to fix the corruption in order to see  
how MySQL will fare when it gets bigger than RAM...