[3223] in linux-scsi channel archive
Re: Mylex 960?
daemon@ATHENA.MIT.EDU (Leonard N. Zubkoff)
Thu Feb 5 22:04:41 1998
Date: Thu, 5 Feb 1998 12:50:38 -0800
From: "Leonard N. Zubkoff" <lnz@dandelion.com>
To: jbrown@tanstaafl.busprod.com
CC: linux-scsi@vger.rutgers.edu
In-reply-to: <Pine.LNX.3.95.980205143150.8350B-100000@shell.busprod.com>
(jbrown@tanstaafl.busprod.com)
Date: Thu, 5 Feb 1998 14:43:26 -0600 (CST)
From: "Justin Brown W." <jbrown@tanstaafl.busprod.com>
Here's the scoop:
Everex Dual PPro 200 w/ 128 megs of ram
Mylex 960 dual channel wide scsi-3 raid only controller
Two Seagate 9.1 gig 7200 rpm barracudas in a raid 0 configuration
Dos 6.0 boot disk
format c /u
Takes 28 seconds by my watch.
Weither the bus can handle it, the controller can handle it, or if the
hand of god reached down, brushed the machine and a miracle occured, it
took 28 seconds. I doubt it's actually reading and writing every byte on
the raid. Based on comparison to ultra-narrow scsi, this thing flew and
that was what shocked our entire technical staff. Our experience with
raid was limited to ultra-narrow DPT up to that point.
You had mentioned something about a full sector test under NT, which I took to
mean that every sector was being tested for readability, which clearly isn't
possible. I can easily believe that a simple DOS format would be very fast,
but I don't believe it accesses all the blocks on a disk.
Leonard
