 |
Direction des applications militaires |  |
 |
Journal articles
MPI detach — Towards automatic asynchronous local completion
Abstract : When aiming for large scale parallel computing, waiting time due to network latency, synchronization, and load imbalance are the primary opponents of high parallel efficiency. A common approach to hide latency with computation is the use of non-blocking communication. In the presence of a consistent load imbalance, synchronization cost is just the visible symptom of the load imbalance. Tasking approaches as in OpenMP, TBB, OmpSs, or C ++20 coroutines promise to expose a higher degree of concurrency, which can be distributed on available execution units and significantly increase load balance. Available MPI non-blocking functionality does not integrate seamlessly into such tasking parallelization. In this work, we present a slim extension of the MPI interface to allow seamless integration of non-blocking communication with available concepts of asynchronous execution in OpenMP and C ++. We furthermore investigate compile-time analysis necessary to transform an application using blocking MPI communication into an application integrating OpenMP tasks with our proposed MPI interface extension.
https://hal-cea.archives-ouvertes.fr/cea-03537990
Contributor : van Man Nguyen Connect in order to contact the contributor
Submitted on : Thursday, January 20, 2022 - 5:01:28 PM
Last modification on : Sunday, January 23, 2022 - 3:24:54 AM
Long-term archiving on: : Thursday, April 21, 2022 - 7:30:29 PM
Contributor : van Man Nguyen Connect in order to contact the contributor
Submitted on : Thursday, January 20, 2022 - 5:01:28 PM
Last modification on : Sunday, January 23, 2022 - 3:24:54 AM
Long-term archiving on: : Thursday, April 21, 2022 - 7:30:29 PM
File
mpi-detach.pdf
Files produced by the author(s)