Building a high-performance collective communication library
Authors:
- 17 December 2002
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 107-116
Abstract
In this paper, we report on a project to develop a unified approach for building a library of collective communication operations that performs well on a cross-section of problems encountered in real applications. The target architecture is a two-dimensional mesh with worm-hole routing, but the techniques are more general. The approach differs from traditional library implementations in that we address the need for implementations that perform well for various sized vectors and grid dimensions, including non-power-of-two grids. We show how a general approach to hybrid algorithms yields performance across the entire range of vector lengths. Moreover, many scalable implementations of application libraries require collective communication within groups of nodes. Our approach yields the same kind of performance for group collective communication. Results from the Intel Paragon system are included. To obtain this library for Intel systems contact intercom©cs.utexas.edu.Keywords
This publication has 8 references indexed in Scilit:
- Global combine on mesh architectures with wormhole routing2002
- Interprocessor collective communication library (InterCom)2002
- The design of a standard message passing interface for distributed memory concurrent computersParallel Computing, 1994
- Broadcasting on linear arrays and meshes1993
- A survey of wormhole routing techniques in direct networksComputer, 1993
- Broadcasting in wraparound meshes with parallel monodirectional linksParallel Computing, 1992
- Efficient Global Combine Operations1991
- Data communication in parallel architecturesParallel Computing, 1989