Return to BSD News archive
Path: sserve!newshost.anu.edu.au!munnari.oz.au!news.Hawaii.Edu!ames!elroy.jpl.nasa.gov!swrinde!cs.utexas.edu!uwm.edu!caen!usenet.coe.montana.edu!bsd.coe.montana.edu!nate From: nate@bsd.coe.montana.edu (Nate Williams) Newsgroups: comp.os.386bsd.questions Subject: Re: NetBSD 0.9, AHA1542C and ST41200N Date: 7 Sep 1993 21:14:16 GMT Organization: Montana Stateu University, Bozeman MT Lines: 47 Message-ID: <26itj8$ebc@pdq.coe.montana.edu> References: <26am3v$det@milou.eunet.fi> NNTP-Posting-Host: bsd.coe.montana.edu In article <26am3v$det@milou.eunet.fi>, Bror 'Count' Heinola <count@muncca.fi> wrote: > (Machine: 386/40 with 8M RAM, Adaptec AHA 1542C SCSI and > Seagate ST41200N 5.25" FH disk) > >5MB file: > >IOZONE performance measurements: > 406424 bytes/second for writing the file > 1024667 bytes/second for reading the file > >10MB file: > >IOZONE performance measurements: > 403557 bytes/second for writing the file > 954697 bytes/second for reading the file > > > Ok, it is not that bad I got earlier (~250k/s write, 600k read) > but it is still not good enough - under DOS (ick!) I got over > 1M/s constantly. 1) Are you running any disk-cache under DOS? 2) The IOZONE benchmarks are troughput through the file-system, of which DOS doesn't have one. :-) The numbers above are very acceptable, and I would say to you that in reality your BSD box will outperform your DOS box in ALL I/O applications. It is difficult to compare DOS benchmarks and *nix benchmarks. To get an idea of how fast BSD can do things, you might try doing this. 'time dd if=/dev/wd0a of=/dev/null' with different block sizes and counts and see what kind of raw speeds you get, which are probably more in-line with what DOS gives you. Nate -- nate@bsd.coe.montana.edu | In the middle of it ........ again. nate@cs.montana.edu | Running/supporting one of many freely available work #: (406) 994-4836 | Operating Systems for [34]86 machines. home #: (406) 586-0579 | (based on Net/2, name changes all the time :-)