In-Reply-To: <200508301132.06904.josh@agliodbs.com>
References: <bcd92b00545ffeef542f68666144099d@medrium.com>
	<200508301132.06904.josh@agliodbs.com>
Mime-Version: 1.0 (Apple Message framework v622)
Content-Type: text/plain; charset=ISO-8859-1; format=flowed
Message-Id: <22924e579ad7e7e0deb5133eb905618b@medrium.com>
Content-Transfer-Encoding: quoted-printable
Cc: pgsql-performance@postgresql.org
From: =?ISO-8859-1?Q?R=E9my_Beaumont?= <remyb@medrium.com>
Subject: Re: High load and iowait but no disk access
Date: Tue, 30 Aug 2005 14:42:38 -0400
To: josh@agliodbs.com


On 30-Aug-05, at 14:32, Josh Berkus wrote:

> Remy,
>
>> The behavior we see is that when running queries that do random reads
>> on disk, IOWAIT goes over 80% and actual disk IO falls to a crawl at =
a
>> throughput bellow 3000kB/s (We usually average 40000 kB/s to 80000=20
>> kB/s
>> on sequential read operations on the netapps)
>
> This seems pretty low for a NetApp -- you should be able to manage up=20=

> to
> 180mb/s, if not higher.   Are you sure it's configured correctly?
Hi Josh,

The config has been reviewed by NetApp. We do get rates higher then=20
80mb/s, but on average, that's what we get.

Do you have NetApp filers deployed ?
How many spindles do you have in your volume ?
On which OS are you running Postgres ?

Thanks,

R=E9my

>
> --=20
> --Josh
>
> Josh Berkus
> Aglio Database Solutions
> San Francisco
>
> ---------------------------(end of=20
> broadcast)---------------------------
> TIP 9: In versions below 8.0, the planner will ignore your desire to
>        choose an index scan if your joining column's datatypes do not
>        match