*BSD News Article 27960


Return to BSD News archive

Path: sserve!newshost.anu.edu.au!munnari.oz.au!bunyip.cc.uq.oz.au!harbinger.cc.monash.edu.au!msuinfo!agate!library.ucla.edu!europa.eng.gtefsd.com!MathWorks.Com!noc.near.net!news.delphi.com!usenet
From: John Dyson <dysonj@delphi.com>
Newsgroups: comp.os.386bsd.development
Subject: Re: Benchmarking VM - another war
Date: Sat, 26 Feb 94 00:20:48 -0500
Organization: Delphi (info@delphi.com email, 800-695-4005 voice)
Lines: 20
Message-ID: <ZkyL+xw.dysonj@delphi.com>
References: <2kgnia$4at@harpo.uccs.edu> <1994Feb24.083301.15046@sunbim.be>
NNTP-Posting-Host: bos1c.delphi.com
X-To: Danny Backx <db@sunbim.be>

Danny Backx <db@sunbim.be> writes:
 
>Making a benchmark to find out which VM system is the best (fastest,
>least or most aggressive, ...) depends on choices. You'll have to
>fight religious battles on what exactly you want to measure and how
>much you will take this feature into account versus the others.
 
I agree.  I can guarantee to you that there will be times that the
NetBSD VM stuff will be faster than the FreeBSD stuff and vice versa.   The
criteria should/might consider other facets of the system too.  Speed is
not the only issue, and even though I know how the VM stuff works, I
honest can't say how to measure ALL of the aspects of the system to say
which is really better in a general sense.  I can still have opinions
and feedback (both positive and negative) from users stating their opinions.
I do have some benchmarks from various sources (including myself) that
I use to measure performance, but wow -- it would be a serious challange
indeed to come up with a VMMARK benchmark :-).
 
John
dyson@implode.root.com