Return to BSD News archive
Path: euryale.cc.adfa.oz.au!newshost.anu.edu.au!harbinger.cc.monash.edu.au!news.rmit.EDU.AU!news.unimelb.EDU.AU!munnari.OZ.AU!spool.mu.edu!howland.reston.ans.net!nntp.crl.com!reason.cdrom.com!usenet From: "Jordan K. Hubbard" <jkh@FreeBSD.org> Newsgroups: comp.os.linux.development.system,comp.unix.bsd.386bsd.misc,comp.unix.bsd.bsdi.misc,comp.unix.bsd.netbsd.misc,comp.unix.bsd.freebsd.misc Subject: The Lai/ Baker paper [was Re: Why to not buy Matrox Millennium] Date: Fri, 05 Apr 1996 11:25:44 -0800 Organization: Walnut Creek CDROM Lines: 38 Message-ID: <316573B8.6201DD56@FreeBSD.org> References: <4jn4qp$6p@darkstar.my.lan> <stephenkDp7nHo.369@netcom.com> <4jrhth$66a@hoopoe.psc.edu> <4jve3t$cfe@hermes.synopsys.com> <4k0m0f$68j@hoopoe.psc.edu> <4k2i4g$n55@usenet.srv.cis.pitt.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 3.0b2 (X11; I; FreeBSD 2.2-CURRENT i386) Xref: euryale.cc.adfa.oz.au comp.os.linux.development.system:20562 comp.unix.bsd.386bsd.misc:426 comp.unix.bsd.bsdi.misc:2959 comp.unix.bsd.netbsd.misc:2725 comp.unix.bsd.freebsd.misc:16543 Mark Hahn wrote: > > > These results were published at the January 1996 USENIX technical conference, > > in a paper by Kevin Lai and Mary Baker from Stanford. I'm sure this > > paper is available on the web -- I apologize for not having a URL, but Look, I attended this presentation and it was a complete travesty - clearly much more of an attempt to get a paper out at USENIX than actually enlighten or make a credible scientific effort. They failed to establish a proper matrix of NFS servers and clients in their tests, chosing instead to use the lowest-performance NFS server they had in all the NFS testing (their excuse: "Oh, well, uh, we only had one PC" GREAT!). They failed to perform the same tests with the same hardware and, to add insult to injury, they closed they presentation by throwing all of the test results out with a statement that "benchmarks don't matter" (great, then why did you just waste 45 minutes of our time showing us your bogus benchmarks?) and that they were chosing Linux because of "the better support they'd received." Not to fail to give Linux full marks for its support, which often is quite good indeed, but these people _never even contacted us for any support_ and so any claims they might have had that Linux's was "better" were not only highly subjective, they were entirely unsubstantiated. I'm sure we all have horror stories to share about being ignored by one group or the other (so let's skip them, OK?) but there are also a number of glowing testimonials about the degree of support provided both by FreeBSD and Linux and I think that the volunteers on both camps do an amazing job - benchmarking "support" is about as meaningful as saying that Linux makes your machine run hotter. This paper was garbage and unfortunate evidence that USENIX's standards for accepting papers are seriously in decline. I'd say this even if they'd picked FreeBSD as "the winner" since conclusions drawn from such sloppy investigation we hardly need on our side! -- - Jordan Hubbard President, FreeBSD Project