Return to BSD News archive
Newsgroups: comp.unix.bsd Path: sserve!manuel.anu.edu.au!munnari.oz.au!news.hawaii.edu!ames!saimiri.primate.wisc.edu!zaphod.mps.ohio-state.edu!menudo.uh.edu!wjin From: wjin@cs.uh.edu (W. Woody Jin) Subject: Re: iozone bogusness Message-ID: <1992Nov10.205929.20974@menudo.uh.edu> Sender: usenet@menudo.uh.edu (USENET News System) Nntp-Posting-Host: modigliani.cs.uh.edu Organization: University of Houston References: <BxE8q0.Czn@agora.rain.com> Date: Tue, 10 Nov 1992 20:59:29 GMT Lines: 25 In article <BxE8q0.Czn@agora.rain.com> davidg@agora.rain.com (David Greenman) writes: >Sorry for the "late" response here, but the first attempt at posting this >never made it through the net. Anyway... > >>I only use iozone 1, just to be consistent and make the test more realistic. >>Most applications do not need more than 1 Megabyte of sequential access.. > > I don't know what you mean by 'realistic'; your results certainly aren't. >Using iozone with a larger value is necessary because, on most machines, the >filesystem cache is at least 1MB and using a test size of less than the cache >size will generate useless results. I have not seen the source codes of io-zone (could anyone point it for me?), but if it writes with O_SYNC mode, and if it reads new files whenever it reads, it certainly does not use any advantage of buffer cache. Otherwise, iozone can give bogus benchmarks. -- ____ ____ ____ ____________________________________ (___) _________________ | | | | | | W. Woody Jin (wjin@cs.uh.edu) (o o) Moo.... | | | |__| | PhD Student. Research Asst. o=======\ / I'm a Cow Lover. | | | | Dept. of Computer Science / | ||O My wife was born \ |---| |--| | University of Houston ` ||'---|| in Cow year. Mooo \____/|__| |__| _______________________________^^ ^^_____________________