DomainKey-Signature: a=rsa-sha1; c=nofws; d=gmail.com; s=gamma;
	h=message-id:date:from:user-agent:mime-version:to:cc:subject
	:references:in-reply-to:content-type:content-transfer-encoding;
	b=ssuuWrGZFIhQqD3r149DCfhOdaKxX6gmPepNXXfpF5dLc6hik5Qq+SKA0EdG0uMjhH
	Xg/8Kh/S5CPL0sA/CAEC+Yx7SjfqX6+4pQ08MswmQbS6gJi8CpeLFdbAJ1mt1PQPWWU+
	Oed6c+vxqpt7CWtwHm4vV7YQilVVuD04MIiss=
Message-ID: <4C4DE50B.1060102@gmail.com>
Date: Mon, 26 Jul 2010 21:42:03 +0200
From: Yeb Havinga <yebhavinga@gmail.com>
User-Agent: Thunderbird 2.0.0.24 (Windows/20100228)
MIME-Version: 1.0
To: Greg Smith <greg@2ndquadrant.com>
CC: pgsql-performance@postgresql.org
Subject: Re: Testing Sandforce SSD
References: <4C4A9452.9070100@gmail.com> <4C4B1F84.5030306@2ndquadrant.com>
	<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>
In-Reply-To: <4C4DAAE8.3060501@2ndquadrant.com>
Content-Type: text/plain; charset=ISO-8859-1; format=flowed
Content-Transfer-Encoding: 7bit

Greg Smith wrote:
> Yeb Havinga wrote:
>> Please remember that particular graphs are from a read/write pgbench 
>> run on a bigger than RAM database that ran for some time (so with 
>> checkpoints), on a *single* $435 50GB drive without BBU raid controller.
>
> To get similar *average* performance results you'd need to put about 4 
> drives and a BBU into a server.  The worst-case latency on that 
> solution is pretty bad though, when a lot of random writes are queued 
> up; I suspect that's where the SSD will look much better.
>
> By the way:  if you want to run a lot more tests in an organized 
> fashion, that's what http://github.com/gregs1104/pgbench-tools was 
> written to do.  That will spit out graphs by client and by scale 
> showing how sensitive the test results are to each.
Got it, running the default config right now.

When you say 'comparable to a small array' - could you give a ballpark 
figure for 'small'?

regards,
Yeb Havinga

PS: Some update on the testing: I did some ext3,ext4,xfs,jfs and also 
ext2 tests on the just-in-memory read/write test. (scale 300) No real 
winners or losers, though ext2 isn't really faster and the manual need 
for fix (y) during boot makes it impractical in its standard 
configuration. I did some poweroff tests with barriers explicitily off 
in ext3, ext4 and xfs, still all recoveries went ok.