[Wien] some more benchmarks

[Peter Blaha]

> We tested :
> * on a cluster of dual 2.4 Ghz xeon machines (2 Gb per node)
> * on a dual 1.8 Ghz opteron (64 bits) machine (4 Gb)

Thank's for this interesting timing. Still a (fairly cheap) P4 (3GHz) is
the fastest machine in this test.

I decided to put this benchmark + timing table on our web-site:

    http://www.wien2k.at/reg_user

Anybody who wants to run this benchmark on a new!!! platform (please do not
repeat benchmarks on "older" or very similar hardware) is very wellcome to
report the results to me.


> * goto is the fastest blas I know (for xeon, at least)

Very interesting result! Intel-PC users should consider to try that library.

> * the opteron is definitely more efficient in handling two jobs at once

That's true, but at the moment (compiler,...) it is still slow.

Regards
                                      P.Blaha
--------------------------------------------------------------------------
Peter BLAHA, Inst.f. Materials Chemistry, TU Vienna, A-1060 Vienna
Phone: +43-1-58801-15671             FAX: +43-1-58801-15698
Email: blaha at theochem.tuwien.ac.at    WWW: http://info.tuwien.ac.at/theochem/
--------------------------------------------------------------------------