Best Point-to-Point Message-Passing Performance
on the IBM SP3-200 (Winterhawk I)
These results represent the highest measured bandwidths for each
of the experiments:
0 to 1 (intranode using US and shared memory MPI)
0 to 2 (internode using US)
0 to 1, 2 to 3 simultaneously (internode using US)
for the single iteration, nonoverlap communication protocols.
Results are presented for bidirectional bandwidth and
unidirectional bandwidth using MPI.
Patrick H. Worley / (
worleyph@ornl.gov)