Hi Lai, On Thu, 2006-09-21 at 00:03, Lai Dragonfly wrote: > Dear all, > > I'm doing the HPL benchmark on several nodes of AMD opteron platform > but may expand to hundred of nodes later. > i hope to get around 80% efficiency. > Anyone has good suggestions of combination of compiler (appreciate if > includes compiler flags) + math lib on AMD platforms?
For HPL, choice of compiler is not too important, but math libraries are very much so. I recommend the Goto BLAS library (You can find it at http://www.tacc.utexas.edu/resources/software/#blas. Expanding to higher numbers of nodes, interconnect can be somewhat important too. You don't need a fantastic interconnect for HPL, but gigE is probably not sufficient for 80% efficiency over hundreds of nodes. -Kevin > thanks a lot. > > PN > > ______________________________________________________________________ > _______________________________________________ > Beowulf mailing list, [email protected] > To change your subscription (digest mode or unsubscribe) visit > http://www.beowulf.org/mailman/listinfo/beowulf _______________________________________________ Beowulf mailing list, [email protected] To change your subscription (digest mode or unsubscribe) visit http://www.beowulf.org/mailman/listinfo/beowulf
