Re: Why Linux, Why Debian
On Sat, 14 Feb 2004 04:44, Matt Zimmerman <mdz@debian.org> wrote:
> On Fri, Feb 13, 2004 at 11:33:17PM +1100, Russell Coker wrote:
> > On Fri, 13 Feb 2004 22:53, Andreas Metzler <ametzler@downhill.at.eu.org>
> >
> > wrote:
> > > http://bulk.fefe.de/scalability/
> > >
> > > Please read all of it, fefe has made number of corrections and
> > > updates.
> >
> > Do you know of something similar that also compares commercial Unices?
>
> He provides the code for his benchmarks, and given their simplicity and the
> fact that they already build and run on the systems tested in the article,
> I imagine they are quite likely to work unchanged on several commercial
> unices.
Yes, all we need to do is get access to such systems...
> Of course, to say that (for example) Solaris is not at its best on x86 was,
> the last time I tried it, a gross understatement.
I think that there are two ways of choosing test hardware. One is to choose
the best of both systems (compare 3GHz P4 or AMD64 CPUs with 900MHz SPARC
CPUs), or compare systems of equivalent price. Either way should give
amusing results.
To compare the kernels we could use CPUs of comparable speed on each side.
400MHz SPARC CPUs seemed to run gzip at the same speed as 400MHz Celeron CPUs
last time I tested.
--
http://www.coker.com.au/selinux/ My NSA Security Enhanced Linux packages
http://www.coker.com.au/bonnie++/ Bonnie++ hard drive benchmark
http://www.coker.com.au/postal/ Postal SMTP/POP benchmark
http://www.coker.com.au/~russell/ My home page
Reply to: