Date: Thu, 29 Jul 2010 10:45:34 -0400
From: Michael Stone <mstone+postgres@mathom.us>
To: pgsql-performance@postgresql.org
Subject: Re: Testing Sandforce SSD
Message-id: <54cd1c80-9b1f-11df-9b6a-001cc0cda50c@msgid.mathom.us>
Mail-followup-to: pgsql-performance@postgresql.org
References: <4C4C83DF.3010703@gmail.com> <4C4C8CCC.7080406@gmail.com>
	<alpine.DEB.2.00.1007261045470.20471@aragorn.flymine.org>
	<4C4D6396.5020707@gmail.com> <4C4DAAE8.3060501@2ndquadrant.com>
	<4C4DE50B.1060102@gmail.com> <4C4DE5C9.8090004@2ndquadrant.com>
	<AANLkTimwAqzvWvSfscqDKJDG3Yod=6KQ6e0Xo+c+k6s=@mail.gmail.com>
	<e7e05556-9a42-11df-9b6a-001cc0cda50c@msgid.mathom.us>
	<4C503473.5080105@gmail.com>
MIME-version: 1.0
Content-type: text/plain; charset=us-ascii; format=flowed
Content-disposition: inline
In-reply-to: <4C503473.5080105@gmail.com>
User-Agent: Mutt/1.5.20 (2009-06-14)

On Wed, Jul 28, 2010 at 03:45:23PM +0200, Yeb Havinga wrote:
>Due to the LBA remapping of the SSD, I'm not sure of putting files 
>that are sequentially written in a different partition (together with 
>e.g. tables) would make a difference: in the end the SSD will have a 
>set new blocks in it's buffer and somehow arrange them into sets of 
>128KB of 256KB writes for the flash chips. See also 
>http://www.anandtech.com/show/2899/2

It's not a question of the hardware side, it's the software. The xlog
needs to by synchronized, and the things the filesystem has to do to 
make that happen penalize the non-xlog disk activity. That's why my 
preferred config is xlog on ext2, rest on xfs. That allows the 
synchronous activity to happen with minimal overhead, while the parts 
that benefit from having more data in flight can do that freely.

Mike Stone