Return to BSD News archive
Path: euryale.cc.adfa.oz.au!newshost.carno.net.au!harbinger.cc.monash.edu.au!munnari.OZ.AU!news.Hawaii.Edu!news.uoregon.edu!arclight.uoregon.edu!news-peer.gsl.net!news.gsl.net!news.mathworks.com!EU.net!news2.EUnet.fr!newsbr.eunet.fr!usenet From: Frederic.Marand@osinet.fr (Frederic MARAND) Newsgroups: comp.os.linux.development.system,comp.unix.bsd.freebsd.misc Subject: Re: Linux vs FreeBSD disk io performance Date: Thu, 17 Oct 1996 06:50:59 GMT Organization: Groupe SEDI / Agorus SA / OSI SARL Lines: 20 Message-ID: <544o8g$rlu@newsbr.eunet.fr> References: <86g23ey1uc.fsf@romulus.ihosteng.priv.no> NNTP-Posting-Host: 193.107.196.155 X-Newsreader: Forte Free Agent 1.0.82 Xref: euryale.cc.adfa.oz.au comp.os.linux.development.system:33604 comp.unix.bsd.freebsd.misc:29415 Ivar Hosteng <ivarh@romulus.ihosteng.priv.no> wrote: >Can someone explain this different behaviour? There's one silly thing I noticed years ago on HP-UX : the code for the /dev/null driver is less efficient than the one for disk I/O. Unbelievable as it seemed to us at the time, we got much better throughput for disk accesses on copies to disk than we got on the same copies to /dev/null. Not having access to the null driver code, we could not see why this happened, but it was really surprising since this does not seem to be a driver with much code in it... Granted, that was years ago, and on HP-UX, not Linux or FreeBSD. But maybe the difference here is also on the /dev/null side, not the disk side. Frederic G. MARAND Agorus SA / OSI SARL Frederic.Marand@osinet.fr