[Beowulf] Benchmarking?

[Roberto Ammendola]

Hi all!

daniel.kidger at quadrics.com wrote:

>Most notably it doesn't really stress MPI, such that most high performance interconnect solutions (IBA, Myrinet, Quadrics et al. [*])  would give fairly similar performance figures.
>

Can somebody quantify an average computation-communication ratio for HPL 
in terms of CPU time?
Thanks

roberto

-- 
______________________________________________________________________

     Roberto Ammendola    INFN - Roma II - APE group
tel: +39-0672594504  email: roberto.ammendola at roma2.infn.it   // \
     Via della Ricerca Scientifica 1 - 00133 Roma             \\_/ //
__________________________________________  ''-.._.-''-.._.. -(||)(')
                                                              '''