Return to BSD News archive
Path: sserve!newshost.anu.edu.au!harbinger.cc.monash.edu.au!bunyip.cc.uq.oz.au!munnari.oz.au!constellation!news.uoknor.edu!ns1.nodak.edu!netnews.nwnet.net!reuter.cse.ogi.edu!hp-cv!hp-pcd!sdd.hp.com!spool.mu.edu!howland.reston.ans.net!math.ohio-state.edu!jussieu.fr!centre.univ-orleans.fr!univ-lyon1.fr!swidir.switch.ch!scsing.switch.ch!news.dfn.de!zeus.rbi.informatik.uni-frankfurt.de!terra.wiwi.uni-frankfurt.de!news.th-darmstadt.de!zib-berlin.de!irz401!uriah!not-for-mail
From: j@uriah.sax.de (J Wunsch)
Newsgroups: comp.os.386bsd.misc
Subject: Re: SCASI or IDE disk?
Date: 19 Aug 1994 14:32:11 +0200
Organization: Private U**X site; member IN e.V.
Lines: 27
Message-ID: <3328obINN4ut@bonnie.sax.de>
References: <salem.136.2E48D0EF@hauk.hsr.no> <CuJL5C.z7M@yuma.acns.colostate.edu> <CuLo9s.4pA@tfs.com> <CunpM0.1J62@yuma.ACNS.ColoState.EDU>
NNTP-Posting-Host: bonnie.sax.de
sandylam@lamar.ColoState.EDU (Sandy Lam) writes:
>: If this is sustained throughput, please let us in on the secret
> as i said, the rates i'm getting is for a program called Coretest
> which measures disk speed. real rates are a lot lower becasue
> of DOS overhead. haven't done teh Unix driver yet, so i can't
>the way i understand coretest, it transmits as much data as it can from the
>drive for 4 seconds. then it's simple division. The numbers aren't exactly
>idyllic, but hell, old IDE drives get maybe 2.5 megs/sec by this test.
HEY, you should better learn about the programs you're using for
``benchmarking'' before posting nonsensish figures here!!!
Coretest is all but a performance test. It reads *the same* block of
data -- up to a maximum of 64 KB -- over and over again. For a modern
IDE disk, it's best suited to measure transfer speeds between the
disk cache and the host, but this number is practically useless.
The data transferred is quickly supplied out of the disk cache, only
a single initial disk read is done to get it from the disk's surface.
Write a U**x driver, and then use iozone on the *raw* disk device.
--
cheers, J"org work: joerg_wunsch@tcd-dresden.de
private: joerg_wunsch@uriah.sax.de
Steinbach's Guideline for Systems Programming:
Never test for an error condition you don't know how to handle.