Message-ID: <4C0AD280.5020105@mtu.net>
Date: Sat, 05 Jun 2010 17:41:04 -0500
From: Jon Schewe <jpschewe@mtu.net>
User-Agent: Mozilla/5.0 (X11; U; Linux x86_64; en-US;
	rv:1.9.1.9) Gecko/20100317 SUSE/3.0.4-1.1.1 Lightning/1.0b1
	Thunderbird/3.0.4
MIME-Version: 1.0
To: Greg Smith <greg@2ndquadrant.com>
CC: pgsql-performance@postgresql.org
Subject: Re: How filesystems matter with PostgreSQL
References: <4C08EEDF.7090403@mtu.net> <4C0AD180.90607@2ndquadrant.com>
In-Reply-To: <4C0AD180.90607@2ndquadrant.com>
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 7bit

On 06/05/2010 05:36 PM, Greg Smith wrote:
> Jon Schewe wrote:
>> The tests were all done on an opensuse 11.2 64-bit machine,
>> on the same hard drive (just ran mkfs between each test) on the same
>> input with the same code base.
>
> So no controller card, just the motherboard and a single hard drive?
Correct.
>   If that's the case, what you've measured is which filesystems are
> safe because they default to flushing drive cache (the ones that take
> around 15 minutes) and which do not (the ones that take >=around 2
> hours).  You can't make ext3 flush the cache correctly no matter what
> you do with barriers, they just don't work on ext3 the way PostgreSQL
> needs them to.
>
So the 15 minute runs are doing it correctly and safely, but the slow
ones are doing the wrong thing? That would imply that ext3 is the safe
one. But your last statement suggests that ext3 is doing the wrong thing.