Date: Tue, 16 Apr 1996 11:23:56 +0200 (MET DST) From: Kees Jan Koster <dutchman@spase.nl> To: freebsd-hackers@FreeBSD.ORG (FreeBSD hackers Mailing list) Message-ID: <199604160923.LAA01450@phobos.spase.nl>
next in thread | raw e-mail | index | archive | help
Hoi hackers,
It's really amusing to see the heated discussions that result when
someone posts the results of a benchmark. Any xxx-ological student
could graduate on this 8)
I figure there are two kinds of benchmarks:
1) configuration tests. Benchmarks to see if a machine performs properly.
2) OS comparisons.
I won't comment on the first type. In the second category I usually see
that the benchmark was run on a few machines.
Wouldn't it be more useful to publish the (source of the) benchmarks and
ask the FreeBSD/netBSD/Linux/SunOS/... community to run it and collect
results (and configuration descriptions) over 1000 or so machines. That
would allow much more solid conclusions. At the same time a larger number
of people are working on the benchmark, eliminating errors and wrong
assumptions.
The results can then be used for configuration testing. This was (is?) done
with the Linux BogoMips. While useless in themselves, there is a list of
configurations, with the expected number of BogoMips.
Scores of people must have run bonnie, why is there no list of disks with
their bonnie results? Someone is bound to use the same cpu and os version
as I do.
Comments?
Groetjes,
Kees Jan
======================================================================v==
Kees Jan Koster e-mail: dutchman@spase.nl
Van Somerenstraat 50 tel: NL-24-3234708
6521 BS Nijmegen
the Netherlands
=========================================================================
Who is this general Failure and why is he reading my disk? (anonymous)
=========================================================================
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199604160923.LAA01450>