Automatically tuned collective communications

  • Authors:

  • Sathish S. Vadhiyar;Graham E. Fagg;Jack Dongarra

  • Affiliations:

  • Department of Computer & Information Science, National Chiao Tung University, Hsinchu, Taiwan 300, R. O. C.;Department of Electrical and Computer Engineering, University of Wisconsin at Madison, Madison, WI;Computer Science and Engineering, Seoul National University, Seoul, 151-742, Korea

  • Venue:
  • Proceedings of the 2000 ACM/IEEE conference on Supercomputing
  • Year:
  • 2000

Quantified Score

Hi-index 0.00

Visualization

Abstract

The performance of the MPI's collective communications is critical in most MPI-based applications. A general algorithm for a given collective communication operation may not give good performance on all systems due to the differences in architectures, network parameters and the storage capacity of the underlying MPI implementation. In this paper we discuss an approach in which the collective communications are tuned for a given system by conducting a series of experiments on the system. We also discuss a dynamic topology method that uses the tuned static topology shape, but re-orders the logical addresses to compensate for changing run time variations. A series of experiments were conducted comparing our tuned collective communication operations to various native vendor MPI implementations. The use of the tuned collective communications resulted in about 30 percent to 650 percent improvement in performance over the native MPI implementations.