Date: Fri, 19 Feb 1999 14:30:32 -0700 (MST) From: "Kenneth D. Merry" <ken@plutotech.com> To: jlemon@americantv.com (Jonathan Lemon) Cc: freebsd-scsi@FreeBSD.ORG Subject: Re: slow disk performance Message-ID: <199902192130.OAA04122@panzer.plutotech.com> In-Reply-To: <19990218215721.55546@right.PCS> from Jonathan Lemon at "Feb 18, 1999 9:57:21 pm"
next in thread | previous in thread | raw e-mail | index | archive | help
Jonathan Lemon wrote...
> I have a system with the following:
>
> ahc0: <Adaptec aic7880 Ultra SCSI adapter> rev 0x01 int a irq 11 on pci0.6.0
> ahc0: aic7880 Wide Channel A, SCSI Id=7, 16/255 SCBs
> da1 at ahc0 bus 0 target 1 lun 0
> da1: <Quantum XP32150W L915> Fixed Direct Access SCSI-2 device
> da1: 20.0MB/s transfers (10.0MHz, offset 8, 16bit), Tagged Queueing Enabled
> da1: 2050MB (4199760 512 byte sectors: 255H 63S/T 261C)
> da0 at ahc0 bus 0 target 0 lun 0
> da0: <IBM DCAS-34330W S65A> Fixed Direct Access SCSI-2 device
> da0: 40.0MB/s transfers (20.0MHz, offset 8, 16bit), Tagged Queueing Enabled
> da0: 4134MB (8467200 512 byte sectors: 255H 63S/T 527C)
>
>
> Basically, an ASUS motherboard with an onboard Adaptec, along with
> two 7200 rpm disks. However, I can't seem to get more than 6MB/sec
> out of the drives, when doing sequential reads or writes:
>
> dd if=/dev/rda1 of=/dev/null bs=64k &
> dd if=/dev/rda0 of=/dev/null bs=64k &
> iostat 1
>
> tty fd0 da0 da1 cpu
> tin tout KB/t tps MB/s KB/t tps MB/s KB/t tps MB/s us ni sy in id
> 0 77 0.00 0 0.00 64.00 115 7.17 64.00 103 6.42 0 0 1 1 98
> 0 76 0.00 0 0.00 64.00 102 6.37 64.00 100 6.25 0 0 1 1 98
> 0 76 0.00 0 0.00 64.00 118 7.36 64.00 103 6.44 0 0 5 0 95
>
>
> Is this the maximum I can expect to get from the drives? I get the same
> speeds when doing sequential writes as well; I would think it can do
> etter than this. There are no "tagged opening" messages in the logfile,
> so this is with 64 tags.
As Steinar Haug pointed out, that may be about all you're going to get from
those disks (he got similar results from his IBM DCAS). The Atlas I was a
decent performer in its day, and I don't think the numbers you're getting
from it are really out of line.
I used to have a 4G Atlas I around, but it died a year and a half ago.
(Quantum was kind enough to replace it with a 9G Atlas II.) But, I do have
a number of Atlas II's kicking around. Running your dd test, I get roughly
9-9.5MB/sec from my two Atlas II's:
fd0 da0 da1 cd0
KB/t tps MB/s KB/t tps MB/s KB/t tps MB/s KB/t tps MB/s
0.00 0 0.00 63.18 150 9.27 64.00 149 9.33 0.00 0 0.00
0.00 0 0.00 63.57 147 9.11 64.00 149 9.29 0.00 0 0.00
0.00 0 0.00 64.00 150 9.35 64.00 148 9.22 0.00 0 0.00
0.00 0 0.00 64.00 149 9.34 64.00 149 9.34 0.00 0 0.00
0.00 0 0.00 64.00 150 9.41 64.00 149 9.28 0.00 0 0.00
0.00 0 0.00 64.00 149 9.28 64.00 149 9.28 0.00 0 0.00
0.00 0 0.00 64.00 149 9.34 64.00 149 9.28 0.00 0 0.00
0.00 0 0.00 64.00 150 9.41 64.00 149 9.28 0.00 0 0.00
0.00 0 0.00 63.58 149 9.27 64.00 146 9.14 0.00 0 0.00
0.00 0 0.00 58.77 131 7.51 64.00 149 9.30 0.00 0 0.00
I get similar results from an Atlas II on another machine that is by itself
on a 7880. I would expect the Atlas I to be somewhat slower for any number
of reasons. (Lower media density being a major culprit.)
Ken
--
Kenneth Merry
ken@plutotech.com
To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe freebsd-scsi" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199902192130.OAA04122>