Abdul Hamid, N.Coddington, P.Vaughan, F.Ouid Khaoua, M.Min, G.2007-07-102007-07-102006Parallel and Distributed Processing Symposium, IPDPS 20th International, 25-29 April, 2006: 8p.1424400546http://hdl.handle.net/2440/36008The results produced by five different MPI benchmark programs on an SGI Altix 3700 are analyzed and compared. There are significant differences in the results for some MPI operations. We investigate the reasons for these discrepancies, which are due to differences in the measurement techniques, implementation details and default configurations of the different benchmarks. The variation in results on the Altix are generally much greater than on a distributed memory machine, due primarily to the ccNUMA architecture and the importance of cache effects, as well as some implementation details of the SGI MPI libraries.enComparison of MPI benchmark programs on an SGI altix ccNUMA shared memory machineConference paper002006267810.1109/IPDPS.2006.16396312-s2.0-3384710695451451Coddington, P. [0000-0003-1336-9686]