*BSD News Article 86792


Return to BSD News archive

Path: euryale.cc.adfa.oz.au!newshost.carno.net.au!harbinger.cc.monash.edu.au!news.cs.su.oz.au!metro!metro!munnari.OZ.AU!news.ecn.uoknor.edu!news.wildstar.net!news.inetnebr.com!newsrelay.courtave.net!polo.iquest.com!news3.crl.com!nexp.crl.com!usenet
From: "Jordan K. Hubbard" <jkh@FreeBSD.org>
Newsgroups: comp.os.linux.misc,comp.os.linux.networking,comp.os.linux.setup,comp.unix.bsd.bsdi.misc,comp.unix.bsd.misc
Subject: Re: Linux vs BSD
Date: Fri, 24 Jan 1997 02:42:13 -0800
Organization: Walnut Creek CDROM
Lines: 37
Message-ID: <32E89205.41C67EA6@FreeBSD.org>
References: <32DFFEAB.7704@usa.net> <5c341j$3dp@cynic.portal.ca> <5c3k6o$qro@lynx.dac.neu.edu> <5c8ifp$445@cynic.portal.ca> <5c8vvb$5uh@lynx.dac.neu.edu>
NNTP-Posting-Host: time.cdrom.com
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
X-Mailer: Mozilla 3.01 (X11; I; FreeBSD 3.0-CURRENT i386)
To: Michael Kagalenko <mkagalen@lynx.dac.neu.edu>
Xref: euryale.cc.adfa.oz.au comp.os.linux.misc:152454 comp.os.linux.networking:65042 comp.os.linux.setup:92257 comp.unix.bsd.bsdi.misc:5605 comp.unix.bsd.misc:1917

Michael Kagalenko wrote:
>
>  In other words, you tried to use NFS performance benchmarks to
>  compare Linux and FreeBSD without knowledge of controller, bus
>  type or network card on the Linux server. That spells
>  "clueless" to me.
> ..
>  It is not, however, possible to make meaningless comparisons between
>  Linux running on PC and FreeBSD on workstations. So your previous
>  post has been idiotic blathering.

Actually, who's the clueless one here?  What do you think FreeBSD runs
on?  PCs.  I don't know what you call a "workstation" here, but FreeBSD
runs on the exact same hardware Linux runs on.  It doesn't run on
SPARCstations or ALPHAs or anything BUT the PC, so it would have been
impossible for the poster to run FreeBSD on "a workstation" as you seem
to be defining it.  If you're going to hurl such charges around, you
should first make sure you're not guilty of the exact same ignorance of
your subject or you only look twice as foolish as the man you're
attacking.

As far as NFS benchmarks are concerned, there have been a number of
independant tests run against both operating systems (and if you're
interested in actually finding out about such tests, you need to go to
conferences like USENIX rather than simply speculating that there is no
hard data available).  Running on the same hardware, NFS performance
will generally be greater on ANY operating system which implements its
NFS services in the kernel rather than in usermode.  That's just one of
life's little trade-offs and Linux appears, for now, to have chosen the
usermode implmementation.  Perhaps there are efforts now underway to
change this and, if so, great.  I'd welcome another university or
research institute sponsored round of benchmarking as I'm sure the
goalposts have moved substantially on both sides since folks like Mary
Baker's group at Stanford ran its last set of comparative benchmarks.
-- 
- Jordan Hubbard
  President, FreeBSD Project