Building a high-performance collective communication library

  • Authors:

  • Mike Barnett;Satya Gupta;David G. Payne;Lance Shuler;Robert van de Geijn;Jerrell Watts

  • Affiliations:

  • University of Idaho, Moscow, Idaho;Supercomputer Systems Division, Intel Corporation, Beaverton, Oregon;Supercomputer Systems Division, Intel Corporation, Beaverton, Oregon;Sandia National Laboratory, Albuquerque, New Mexico;The University of Texas at Austin, Austin, Texas;California Institute of Technology, Pasadena, California

  • Venue:
  • Proceedings of the 1994 ACM/IEEE conference on Supercomputing
  • Year:
  • 1994

Quantified Score

Hi-index 0.00

Visualization

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.