TY - GEN
T1 - A scalable MPI-Comm-split algorithm for exascale computing
AU - Sack, Paul
AU - Gropp, William D
PY - 2010
Y1 - 2010
N2 - Existing algorithms for creating communicators in MPI programs will not scale well to future exascale supercomputers containing millions of cores. In this work, we present a novel communicator-creation algorithm that does scale well into millions of processes using three techniques: replacing the sorting at the end of MPI-Comm-split with merging as the color and key table is built, sorting the color and key table in parallel, and using a distributed table to store the output communicator data rather than a replicated table. This reduces the time cost of MPI-Comm-split in the worst case we consider from 22 seconds to 0.37 second. Existing algorithms build a table with as many entries as processes, using vast amounts of memory. Our algorithm uses a small, fixed amount of memory per communicator after MPI-Comm-split has finished and uses a fraction of the memory used by the conventional algorithm for temporary storage during the execution of MPI-Comm-split.
AB - Existing algorithms for creating communicators in MPI programs will not scale well to future exascale supercomputers containing millions of cores. In this work, we present a novel communicator-creation algorithm that does scale well into millions of processes using three techniques: replacing the sorting at the end of MPI-Comm-split with merging as the color and key table is built, sorting the color and key table in parallel, and using a distributed table to store the output communicator data rather than a replicated table. This reduces the time cost of MPI-Comm-split in the worst case we consider from 22 seconds to 0.37 second. Existing algorithms build a table with as many entries as processes, using vast amounts of memory. Our algorithm uses a small, fixed amount of memory per communicator after MPI-Comm-split has finished and uses a fraction of the memory used by the conventional algorithm for temporary storage during the execution of MPI-Comm-split.
UR - http://www.scopus.com/inward/record.url?scp=78149279221&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=78149279221&partnerID=8YFLogxK
U2 - 10.1007/978-3-642-15646-5_1
DO - 10.1007/978-3-642-15646-5_1
M3 - Conference contribution
AN - SCOPUS:78149279221
SN - 3642156452
SN - 9783642156458
T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
SP - 1
EP - 10
BT - Recent Advances in the Message Passing Interface - 17th European MPI Users' Group Meeting, EuroMPI 2010, Proceedings
T2 - 17th European MPI Users' Group Meeting, EuroMPI 2010
Y2 - 12 September 2010 through 15 September 2010
ER -