*BSD News Article 7501


Return to BSD News archive

Xref: sserve comp.arch:27931 comp.unix.bsd:7550 comp.os.linux:14998
Newsgroups: comp.arch,comp.unix.bsd,comp.os.linux
Path: sserve!manuel.anu.edu.au!munnari.oz.au!sgiblab!zaphod.mps.ohio-state.edu!cs.utexas.edu!uunet!walter!gizmo!mo
From: mo@gizmo.bellcore.com (Michael O'Dell)
Subject: Re: IDE faster than Mips SCSI disk?????
Message-ID: <1992Nov6.142946.17430@walter.bellcore.com>
Keywords: 
 
Sender: news@walter.bellcore.com
Nntp-Posting-Host: gizmo.bellcore.com
Reply-To: mo@bellcore.com
Organization: Center for Chaotic Repeatabilty
References: <1992Nov6.033942.21194@ntuix.ntu.ac.sg>
Date: Fri, 6 Nov 92 14:29:46 GMT
Lines: 27

from our experiences trying to qualify drives and controllers
for a computer project i was involved with a few years ago....

one must be very careful that one isn't comparing the
performance of two disk controllers and not the attached drives.
it is VERY tricky to sort out why one disk goes faster
than another, and when and under what circumstances.
controller startup latencies vary by geologic timescales,
or so it seems if one tries hard to measure them accurately.
we'll not even go into latencies caused by the number of
times the controller must be touched to actually start a transfer.

the same caveats apply to measuring the performance
of the underlying drive and electronics (with it's
embedded controller <see above>) versus
measuring the performance of the transfer channel
implementation.

transfer channel protocols (IDE vs SCSI-2) are compared
with timing diagrams, but that is admittedly less interesting
than comparing implementations which govern what one
really gets.  the problem, however, is really *knowing*
what one is measuring.

	-Mike O'Dell

Bellcore??? Bellcore isn't allowed opinions. Any found here are mine.