DMCA
High Performance Implementation of MPI Derived Datatype Communication over InfiniBand (2004)
| Citations: | 4 - 0 self |
BibTeX
@MISC{Wu04highperformance,
author = {Jiesheng Wu and Pete Wyckoff and Dhabaleswar Panda},
title = {High Performance Implementation of MPI Derived Datatype Communication over InfiniBand},
year = {2004}
}
Share
 |
 |
 |
 |
||
OpenURL
Abstract
In this paper, a systematic study of two main types of approach for MPI datatype communication (Pack/Unpack- based approaches and Copy-Reduced approaches)iscar- ried out on the InfiniBand network. We focus on overlapping packing, network communication, and unpacking in the Pack/Unpack-based approaches. We use RDMA operations to avoid packing and/or unpacking in the CopyReduced approaches. Four schemes (Buffer-Centric Segment Pack/Unpack, RDMA Write Gather With Unpack, Pack with RDMA Read Scatter, and Multiple RDMA Writes have been proposed. Three of them have been implemented and evaluated based on one MPI implementation over InfiniBand. Performance results of a vector microbenchmark demonstrate that latency is improved by a factor of up to 3.4 and bandwidth by a factor of up to 3.6 compared to the current datatype communication implementation. Collective operations like MPI Alltoall are demonstrated to benefit. A factor of up to 2.0 improvement has been seen in our measurements of those collective operations on an 8-node system.
Keyphrases
mpi derived datatype communication high performance implementation collective operation infiniband network network communication current datatype communication implementation mpi alltoall buffer-centric segment pack unpack pack unpack-based approach copyreduced approach rdma read scatter rdma write gather performance result 8-node system mpi datatype communication pack unpack vector microbenchmark demonstrate multiple rdma writes main type rdma operation systematic study mpi implementation copy-reduced approach
