Date: Mon, 13 Sep 2004 15:47:49 -0400 From: Bill Moran <wmoran@potentialtech.com> To: Stefan Cars <stefan@snowfall.se> Cc: questions@freebsd.org Subject: Re: Sysadmin magazine benchmarkings Message-ID: <20040913154749.184a38c7.wmoran@potentialtech.com> In-Reply-To: <20040913211740.Y71085@pluring.snowfall.se> References: <20040913211740.Y71085@pluring.snowfall.se>
next in thread | previous in thread | raw e-mail | index | archive | help
Stefan Cars <stefan@snowfall.se> wrote: > Hi! > > I read the article on sys admin magazine > (http://www.samag.com/documents/s=1148/sam0107a/0107a.htm) where they > benchmarked Linux, Windows, Solaris and FreeBSD regarding High-Performance > Network Applications. This article is from 2001 and covers FreeBSD 4.2. > Anyhow, FreeBSD ended last which feels quiet strange, is there anyone that > has any new benchmarking data on how FreeBSD 5 compares ? I wouldn't place any validity, whatsoever, on those tests. They claim that FreeBSD is 6x slower on disk writes than Linux and Windows. I say hogwash: http://www.potentialtech.com/wmoran/postgresql.php#results You can see that FreeBSD is only slightly slower than Linux in these tests ... I find it hard to believe that a properly configured FreeBSD system could ever be 6x slower than Linux. I find it much easier to believe that the people who ran the tests don't know how to set up FreeBSD. Do some searches and you'll find that this artical has been discussed on this list ad-nausium ... since it's so obviously bogus or biased. They don't even give details of the hardware or software setups used to test, so it's possible that the HDD controller was not fully supported by FreeBSD, or that the disks were mounted full-sync. -- Bill Moran Potential Technologies http://www.potentialtech.com
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20040913154749.184a38c7.wmoran>