[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.
Greg Lindahl lindahl at pbm.comThu Apr 17 10:43:05 PDT 2008
- Previous message: [Beowulf] OpenMP vs. MPI benchmark for multi-core machines?
- Next message: [Beowulf] OpenMP vs. MPI benchmark for multi-core machines?
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
On Thu, Apr 17, 2008 at 03:38:41PM +0300, Eray Ozkural wrote: > 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). I can point to several examples of codes which have both an MPI and OpenMP version, where the MPI code runs faster. Locality is your friend. Unfortunately, the most obvious set to compare (NAS Parallel Benchmarks) uses different serial algorithms for the OpenMP and MPI codes :-/ This topic has come up several times on this list in the past year. -- greg
- Previous message: [Beowulf] OpenMP vs. MPI benchmark for multi-core machines?
- Next message: [Beowulf] OpenMP vs. MPI benchmark for multi-core machines?
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
More information about the Beowulf mailing list
