[Beowulf] OpenMP vs. MPI benchmark for multi-core machines?
Many of your questions may have already been answered in earlier discussions or in the FAQ. The search results page will indicate current discussions as well as past list serves, articles, and papers.
Douglas Eadline deadline at eadline.orgThu Apr 17 13:21:51 PDT 2008
- Previous message: [Beowulf] OpenMP vs. MPI benchmark for multi-core machines?
- Next message: [Beowulf] tape backup software for linux server with windows and linux clients.
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
you may want to look at this: http://www.linux-mag.com/id/4608 (registration required) NAS parallel OpenMP vs NAS Parallel MPI on clovertown node (dual socket, 8 cores total). As Greg said, the programs are different but, I suppose it is better than nothing. I should mention that such comparisons are hard because a good shared memory version of an algorithm and an MPI version of the same problem are usually very different. If you can find a copy, Pfister's "In Search of Clusters" devotes a whole chapter to this topic (in detail). -- Doug > Hi there, > > Is there such a benchmark that I can refer to? I am increasingly > convinced that OpenMP/pthread is required only in extreme cases, but I > need some numbers to prove that (to myself and my advisor). > > Best, > > -- > Eray Ozkural, PhD candidate. Comp. Sci. Dept., Bilkent University, Ankara > _______________________________________________ > Beowulf mailing list, Beowulf at beowulf.org > To change your subscription (digest mode or unsubscribe) visit > http://www.beowulf.org/mailman/listinfo/beowulf > > !DSPAM:48078b7e183188298414181! > -- Doug
- Previous message: [Beowulf] OpenMP vs. MPI benchmark for multi-core machines?
- Next message: [Beowulf] tape backup software for linux server with windows and linux clients.
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
More information about the Beowulf mailing list
