136
click to vote
PDP
2009
IEEE
15 years 9 months ago
2009
IEEE
Abstract—The increasing number of cores per node in highperformance computing requires an efficient intra-node MPI communication subsystem. Most existing MPI implementations rel...