Return to BSD News archive
Path: euryale.cc.adfa.oz.au!newshost.carno.net.au!harbinger.cc.monash.edu.au!munnari.OZ.AU!news.ecn.uoknor.edu!feed1.news.erols.com!howland.erols.net!math.ohio-state.edu!news.cis.ohio-state.edu!nntp.sei.cmu.edu!news.psc.edu!not-for-mail From: peterb@hoopoe.psc.edu (Peter Berger) Newsgroups: comp.unix.bsd.freebsd.misc,comp.unix.bsd.bsdi.misc,comp.sys.sgi.misc Subject: Re: no such thing as a "general user community" Date: 30 Mar 1997 22:22:49 -0500 Organization: Pittsburgh Supercomputing Center Lines: 62 Message-ID: <5hnam9$393@hoopoe.psc.edu> References: <331BB7DD.28EC@net5.net> <333EA3EF.41C67EA6@consys.com> <333EE416.ABD322C@FreeBSD.org> <5hn00k$dio@fido.asd.sgi.com> NNTP-Posting-Host: hoopoe.psc.edu Xref: euryale.cc.adfa.oz.au comp.unix.bsd.freebsd.misc:38110 comp.unix.bsd.bsdi.misc:6519 comp.sys.sgi.misc:29545 In article <5hn00k$dio@fido.asd.sgi.com>, Larry McVoy <lm@slovax.engr.sgi.com> wrote: >Jordan K. Hubbard (jkh@FreeBSD.org) wrote: >: Still, since Larry has now proven to us all that one can blatently get >: away with this kind of sleight-of-hand marketing > >Jordan, your unhappiness with that Usenix paper has nothing to do with the >numbers. I just went through the numbers, which were certainly apples to >apples, slightly skewed in favor of FreeBSD (133Mhz P5 vs 120Mhz P5 for >Linux). Here's the summary of the i586 systems, FreeBSD vs Linux: > >FreeBSD wins: memory bandwidth (this is really hardware, not OS), TCP > bandwidth (local & remote), file reread (read & mmap), signal > handling, TCP latency (local & remote), TCP connect latency. >Linux wins: pipe bw, system call, fork, exec, shell, contex switching, pipe > pipe latency, udp latency, file system create / delete. Larry, this sort of nonsense is completely beneath you. I think less and less of you every time you repeat it. The point is that it doesn't matter which way -you- think the hardware differences are skewed. Frankly, I'm sure that Jordan would have objected to your conclusions even if FreeBSD won across the board: because your conclusions are completely, utterly, and didactically bogus. You can't compare apples to oranges and then claim that the oranges are apples because you like apple juice. The point is that saying "FreeBSD wins" or "Linux wins" when the underlying hardware you're making your "comparisons" on is different. If I compared two versions of IRIX, one running on an Indy and another running on a Personal Iris and then told the world that "IRIX X.X wins: pipe bw, system call, etc." the world would ask me what sort of crack I was smoking. And they would be right. Testing FreeBSD versus Linux on the same hardware platform is truly trivial -- if you're interested in actually doing the work. You indicated, at that talk, that you were not interested in doing that work. That's fine. But the community is perfectly justified in pointing out that drawing any sort of useful comparisons from said data is impossible, especially when it is clear that *BSD and Linux suffer different performance hits when starved for different resources. You can run the same OS on different hardware, and use your benchmarks to make statements about the comparative value of the hardware. You can run different OSes on the same hardware, and use your benchmarks to make statements about the comparative value of the OSs. But running different OSes on different hardware only allows us to make statements about the comparative lack of rigor on the part of the researcher. Sorry, but I call 'em as I see 'em. -- Pete Berger, Esq. Coordinator, Regional Information Infrastructure Pittsburgh Supercomputing Center peterb@psc.edu http://www.psc.edu/~peterb I don't speak for my employers, nor they for me.