Path: utzoo!mnetor!tmsoft!torsqnt!lethe!yunexus!ists!helios.physics.utoronto.ca!news-server.csri.toronto.edu!cs.utexas.edu!usc!chaph.usc.edu!aludra.usc.edu!karrer
From: karrer@aludra.usc.edu (Anthony Karrer)
Newsgroups: comp.os.minix
Subject: File System Performance Benchmark
Message-ID: <14759@chaph.usc.edu>
Date: 7 Feb 91 20:41:04 GMT
References: <1991Jan30.173551.4547@midway.uchicago.edu> <910202513@uwalt.nl.mugnet.org> <34280@ditka.Chicago.COM>
Sender: news@chaph.usc.edu
Organization: University of Southern California, Los Angeles, CA
Lines: 13
Nntp-Posting-Host: aludra.usc.edu

One of the assignments I would like the students to do this is
minimal attempts at improving the performance of the file system.
However, when I tried to create a benchmark which they could use
in evaluating their solutions, I found some problems.  The first
problem is that the clock rate (at least the only one I know of)
is set to 60Hz so that a performance improvement on the order of
.1 ms/access would be unnoticable except in extremely large
tests.  Bottom line ... How can I create a semi-reasonable file
system performance benchmark?

-- 
Take It Easy,
  Anthony Karrer (akarrer@pollux.usc.edu)