Skip site navigation (1)Skip section navigation (2)
Date:      Sun, 14 Sep 1997 19:04:37 +0000
From:      Josh Tiefenbach <josh@doun.org>
To:        Stefan Esser <se@freebsd.org>
Cc:        hackers@freebsd.org, josh@doun.org
Subject:   Re: Problems with NCR810/Micropolis disk.
Message-ID:  <19970914190437.02165@doun.org>
In-Reply-To: <19970914233827.50442@mi.uni-koeln.de>; from Stefan Esser on Sun, Sep 14, 1997 at 11:38:27PM %2B0200
References:  <19970913204749.41765@doun.org> <19970914233827.50442@mi.uni-koeln.de>

next in thread | previous in thread | raw e-mail | index | archive | help
> > In searching the archives, the I found a post to -scsi circa 12/96 with the
> > exact same dmesg output (M_DISCONNEXT) involving an ncr810 and a 9GB
> > Micropolis drive, but no solution/explanation was offered.
> 
> Really ? I don't remember having seen such a message ...
> Will have to check the archives myself ...

Search on -scsi for Joe Greco's post on Dec 10/96

> 
> > Could this concievably be a hardware problem (ie, should I return the drive),
> > or is there something patently obvious that I'm missing?
> 
> Well, I don't know how you measured those 250KB/s 
> numbers. Could you please run bonnie (from ports) or

Bonnie results:

				 -------Sequential Output-------- ---Sequential Input-- --Random--
              -Per Char- --Block--- -Rewrite-- -Per Char- --Block--- --Seeks---
Machine    MB K/sec %CPU K/sec %CPU K/sec %CPU K/sec %CPU K/sec %CPU  /sec %CPU
          100  4046 49.4  4104 10.1  1675  6.8  4374 57.2  4661 13.4  87.5 2.8

> send results of "dd" with block sizes of 512 byte,
> 4KB and 16KB ?

dd results (in all cases, if=/dev/rsd0a of=/dev/null count = 10000)
asherah:~# dd if=/dev/rsd0a of=/dev/null bs=512 count=10000
10000+0 records in
10000+0 records out
5120000 bytes transferred in 19.132326 secs (267610 bytes/sec)

asherah:~# dd if=/dev/rsd0a of=/dev/null bs=4k count=10000
8192+0 records in
8192+0 records out
33554432 bytes transferred in 18.025235 secs (1861525 bytes/sec)

asherah:~# dd if=/dev/rsd0a of=/dev/null bs=16k count=10000
2048+0 records in
2048+0 records out
33554432 bytes transferred in 6.669969 secs (5030673 bytes/sec)

Hmmm. I see that with 16k blocksizes, the performance is much better, but the
drive still feels `pokey'. 

-- 
HARANGUE, n.
  A speech by an opponent, who is known as an harrangue-outang

                       -- Ambrose Bierce



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?19970914190437.02165>