Sciweavers

Share
ICS
2005
Tsinghua U.

Optimization of MPI collective communication on BlueGene/L systems

9 years 2 months ago
Optimization of MPI collective communication on BlueGene/L systems
BlueGene/L is currently the world’s fastest supercomputer. It consists of a large number of low power dual-processor compute nodes interconnected by high speed torus and collective networks. Because compute nodes do not have shared memory, MPI is the the natural programming model for this machine. The BlueGene/L MPI library is a port of MPICH2. In this paper we discuss the implementation of MPI collectives on BlueGene/L. The MPICH2 implementation of MPI collectives is based on point-to-point communication primitives. This turns out to be suboptimal for a number of reasons. Machine-optimized MPI collectives are necessary to harness the performance of BlueGene/L. We discuss these optimized MPI collectives, describing the algorithms and presenting performance results measured with targeted micro-benchmarks on real BlueGene/L hardware with up to 4096 compute nodes. Categories and Subject Descriptors C.4 [Computer Systems Organization]: Performance
George Almási, Philip Heidelberger, Charles
Added 27 Jun 2010
Updated 27 Jun 2010
Type Conference
Year 2005
Where ICS
Authors George Almási, Philip Heidelberger, Charles Archer, Xavier Martorell, C. Christopher Erway, José E. Moreira, Burkhard D. Steinmacher-Burow, Yili Zheng
Comments (0)
books