Performance evaluation of Allgather algorithms on terascale Linux cluster with fast Ethernet

Author

Chen, Jing ; Zhang, Yunquan ; Zhang, Linbo ; Yuan, Wei

Author_Institution

Dept. of Comput. Sci., China Univ. of Sci. & Technol.

fYear

2005

fDate

1-1 July 2005

Lastpage

442

Abstract

We report our work on evaluating performance of several MPI Allgather algorithms on fast Ethernet. These algorithms are ring, recursive doubling, Bruck, and neighbor exchange. The first three algorithms are widely used today. The neighbor exchange algorithm which was proposed by the authors incorporates pair-wise exchange, and is expected to perform better with certain configurations, mainly when using TCP/IP over Ethernet. We tested the four algorithms on terascale Linux clusters DeepComp 6800 and DAWNING 4000A using TCP/IP over fast Ethernet. Results show that our neighbor exchange algorithm performs the best for long messages, the ring algorithm performs the best for medium-size messages and the recursive doubling algorithm performs the best for short messages

Keywords

message passing; performance evaluation; transport protocols; workstation clusters; Bruck algorithm; DAWNING 4000A; DeepComp 6800; MPI Allgather algorithm; TCP/IP; fast Ethernet; neighbor exchange algorithm; pair-wise exchange; performance evaluation; recursive doubling algorithm; ring algorithm; terascale Linux cluster; Clustering algorithms; Computer science; Content addressable storage; Costs; Ethernet networks; High performance computing; Linux; Parallel machines; TCPIP; Testing;

fLanguage

English

Publisher

ieee

Conference_Titel

High-Performance Computing in Asia-Pacific Region, 2005. Proceedings. Eighth International Conference on

Conference_Location

Beijing

Print_ISBN

0-7695-2486-9

Type

conf

DOI

10.1109/HPCASIA.2005.75

Filename

1592302