Skip site navigation (1)Skip section navigation (2)
Date:      Thu, 2 Mar 2006 08:41:33 -0600
From:      "Nikolas Britton" <nikolas.britton@gmail.com>
To:        "Alex Zbyslaw" <xfb52@dial.pipex.com>
Cc:        Beastie <beastie@mra.co.id>, Liste FreeBSD <freebsd-questions@freebsd.org>
Subject:   Re: SATA Raid (stress test..)
Message-ID:  <ef10de9a0603020641t7014bf4cn9c9cc08b8d62af29@mail.gmail.com>
In-Reply-To: <4406CB4D.5050300@dial.pipex.com>
References:  <61560.207.70.139.52.1139628926.squirrel@www.compedgeracing.com> <65260.207.70.139.52.1139998857.squirrel@www.compedgeracing.com> <06b901c63220$3a849eb0$c801a8c0@nexpc> <50778.207.70.139.52.1140002253.squirrel@www.compedgeracing.com> <43F3EDD6.80707@mra.co.id> <44052663.7000802@mra.co.id> <440565FF.3030002@mra.co.id> <44058D9E.3010801@dial.pipex.com> <440675E0.1020204@mra.co.id> <4406CB4D.5050300@dial.pipex.com>

next in thread | previous in thread | raw e-mail | index | archive | help
On 3/2/06, Alex Zbyslaw <xfb52@dial.pipex.com> wrote:
>
[snipped]
>
> Why not happy?  Transfer rates from 53 to 92Mb/s, give or take; what's
> wrong with that?  On a plain sata disk I get:
>
> Seek times:
>         Full stroke:      250 iter in   4.717248 sec =3D   18.869 msec
>         Half stroke:      250 iter in   5.342099 sec =3D   21.368 msec
>         Quarter stroke:   500 iter in   8.870424 sec =3D   17.741 msec
>         Short forward:    400 iter in   2.753187 sec =3D    6.883 msec
>         Short backward:   400 iter in   1.390941 sec =3D    3.477 msec
>         Seq outer:       2048 iter in   0.426796 sec =3D    0.208 msec
>         Seq inner:       2048 iter in   0.487280 sec =3D    0.238 msec
> Transfer rates:
>         outside:       102400 kbytes in   1.652736 sec =3D    61958 kbyte=
s/sec
>         middle:        102400 kbytes in   1.697364 sec =3D    60329 kbyte=
s/sec
>         inside:        102400 kbytes in   1.834759 sec =3D    55811 kbyte=
s/sec
>
> A second, different, disk gives me better seek times but roughly similar
> transfer rates.  So I beat your inside transfer rate, but you're 50% up
> on the outside rate.
>
> If you have windows anywhere, then download sandra-lite.  Among other
> things, it has comparison benchmarks for all its tests, including disk
> transfer rates for things like SCSI-RAID0, RAID1, SATA/PATA-RAID0/1 etc.
>

diskinfo -t /dev/da0e
/dev/da0
        512             # sectorsize
        1756440297472   # mediasize in bytes (1.6T)
        3430547456      # mediasize in sectors
        213541          # Cylinders according to firmware.
        255             # Heads according to firmware.
        63              # Sectors according to firmware.

Seek times:
        Full stroke:      250 iter in   3.502539 sec =3D   14.010 msec
        Half stroke:      250 iter in   2.749807 sec =3D   10.999 msec
        Quarter stroke:   500 iter in   4.919431 sec =3D    9.839 msec
        Short forward:    400 iter in   2.257898 sec =3D    5.645 msec
        Short backward:   400 iter in   2.293915 sec =3D    5.735 msec
        Seq outer:       2048 iter in   0.132233 sec =3D    0.065 msec
        Seq inner:       2048 iter in   0.152473 sec =3D    0.074 msec
Transfer rates:
        outside:       102400 kbytes in   1.487112 sec =3D    68858 kbytes/=
sec
        middle:        102400 kbytes in   1.505039 sec =3D    68038 kbytes/=
sec
        inside:        102400 kbytes in   1.336495 sec =3D    76618 kbytes/=
sec


This and all the other benchmarks you've run are useless. Run a real
benchmark like iozone. It's in ports under benchmarks/iozone.
http://www.iozone.org/



--
BSD Podcasts @ http://bsdtalk.blogspot.com/



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?ef10de9a0603020641t7014bf4cn9c9cc08b8d62af29>