Performance Comparison of MPI Implementations over InfiniBand, Myrinet and Quadrics

Author

Liu, Jiuxing ; Chandrasekaran, Balasubramanian ; Wu, Jiesheng ; Jiang, Weihang ; Kini, Sushmitha ; Yu, Weikuan ; Buntinas, Darius ; Wyckoff, Peter ; Panda, D.K.

Author_Institution

The Ohio State University, Columbus

fYear

2003

fDate

15-21 Nov. 2003

Firstpage

58

Lastpage

58

Abstract

In this paper, we present a comprehensive performance comparison of MPI implementations over Infini-Band, Myrinet and Quadrics. Our performance evaluation consists of two major parts. The first part consists of a set of MPI level micro-benchmarks that characterize different aspects of MPI implementations. The second part of the performance evaluation consists of application level benchmarks. We have used the NAS Parallel Benchmarks and the sweep3D benchmark. We not only present the overall performance results, but also relate application communication characteristics to the information we acquired from the micro-benchmarks. Our results show that the three MPI implementations all have their advantages and disadvantages. For our 8-node cluster, InfiniBand can offer significant performance improvements for a number of applications compared with Myrinet and Quadrics when using the PCI-X bus. Even with just the PCI bus, InfiniBand can still perform better if the applications are bandwidth-bound.

Keywords

Bandwidth; Contracts; Delay; Distributed computing; High performance computing; Information science; Laboratories; Permission; Personal communication networks; Supercomputers;

fLanguage

English

Publisher

ieee

Conference_Titel

Supercomputing, 2003 ACM/IEEE Conference

Print_ISBN

1-58113-695-1

Type

conf

DOI

10.1109/SC.2003.10007

Filename

1592961