[Beowulf] Really efficient MPIs??
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.
f.costen at cs.man.ac.uk fumie.costen at manchester.ac.ukWed Nov 28 07:57:37 PST 2007
- Previous message: [Beowulf] Really efficient MPIs??
- Next message: [Beowulf] Really efficient MPIs??
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
Dear Jeff > If you are going to use GigE (TCP), I would recommend Scali MPI > (www.scali.com). It's commercial, but it's the best MPI I've ever tested > (the fastest). Plus it works with TCP, Myrinet, and IB without having > to recompile. > > If you don't want to pay money for an MPI, then go with Open-MPI. So Open-MPI is different from OpenMP, isn't it ? as OpenMP is only for the shared memory architecture. I did various tests for the hybrid case( OpenMP and MPI ) for my FDTD(finite difference time domain) calculation and when I use supercomputers[particularly Bull](using 8 nodes and 8 cores in one node), calculation with MPI for all the communication is faster than OpenMP. Best wishes, Fumie
- Previous message: [Beowulf] Really efficient MPIs??
- Next message: [Beowulf] Really efficient MPIs??
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
More information about the Beowulf mailing list
