Skip site navigation (1)Skip section navigation (2)
Date:      Thu, 25 Apr 2002 13:27:47 +0930
From:      Greg 'groggy' Lehey <grog@FreeBSD.org>
To:        Mike Tancsa <mike@sentex.net>
Cc:        stable@FreeBSD.ORG
Subject:   Re: new ata code and vinum
Message-ID:  <20020425132747.G33950@wantadilla.lemis.com>
In-Reply-To: <5.1.0.14.0.20020422210120.05d72808@192.168.0.12>
References:  <5.1.0.14.0.20020422160043.03499a50@marble.sentex.ca> <5.1.0.14.0.20020422160043.03499a50@marble.sentex.ca> <5.1.0.14.0.20020422210120.05d72808@192.168.0.12>

next in thread | previous in thread | raw e-mail | index | archive | help
On Monday, 22 April 2002 at 21:06:08 -0400, Mike Tancsa wrote:
> At 09:29 AM 4/23/2002 +0930, Greg 'groggy' Lehey wrote:
>> [Format recovered--see http://www.lemis.com/email/email-format.html]
>>
>> bonnie tests mainly the CPU, not the storage.  The only test of any
>> interest for the storage system is the random seeks.  Use rawio if you
>> want too get some useful figures.
>
> Here you go! vinum0 is the SCSI pair, vinum1 the 2 seagate IDEs.
>
>            Random read  Sequential read    Random write Sequential write
> ID          K/sec  /sec    K/sec  /sec     K/sec  /sec     K/sec  /sec
> vinum0     3953.4   244   5219.1   319    4646.6   289    5430.8   331
>            Random read  Sequential read    Random write Sequential write
> ID          K/sec  /sec    K/sec  /sec     K/sec  /sec     K/sec  /sec
> vinum1     2038.6   126  17370.4  1060    2985.2   185   44608.7  2723
>            Random read  Sequential read    Random write Sequential write
> ID          K/sec  /sec    K/sec  /sec     K/sec  /sec     K/sec  /sec
> ar0s1e     1802.0   112  11960.7   730    2628.6   165   29238.0  1785
>            Random read  Sequential read    Random write Sequential write
> ID          K/sec  /sec    K/sec  /sec     K/sec  /sec     K/sec  /sec
> da0s1e     2550.5   157   2772.6   169    2424.4   150    2802.9   171
>            Random read  Sequential read    Random write Sequential write
> ID          K/sec  /sec    K/sec  /sec     K/sec  /sec     K/sec  /sec
> ad10s1e    1248.6    77   2242.6   137    1645.7   102    7207.9   440

Yes, that's more like what I would expect.  It's interesting to note
that the ar doesn't perform as well as Vinum.

> I was going to use the pair of SCSIs for the news database... The other
> drives for the articles.

Looks like you have write cacheing turned off on the SCSI drives.  Is
that intentional?

Greg
--
See complete headers for address and phone numbers

To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe freebsd-stable" in the body of the message




Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20020425132747.G33950>