A selection theory and methodology for heterogeneous supercomputing

Document Type

Conference Proceeding

Publication Date

1-1-1993

Abstract

In this paper, a methodology for mapping algorithms onto heterogeneous suite of supercomputers is presented. An approach for selecting an optimal suite of computers for solving problems with diverse computational requirements, called Heterogeneous Optimal Selection Theory (HOST), is presented. HOST is an extension to Augmented Optimal Selection Theory in two ways: It incorporates heterogeneous parallelism embedded in the tasks, and it reflects the costs associated in using various fine grain mapping strategies at individual machine level. The proposed mapping methodology is based on the Cluster-M programming paradigm. For the mapping purpose, the input format, assumed in HOST, is modeled in terms vf Hierarchical Cluster-M specification and representation. For a given problem, a Hierarchical Cluster-M specification is generated to indicate the execution of concurrent tasks at different stages of the computation. This specification is then mapped onto the Hierarchical Cluster-M representation of the underlying heterogeneous suite of supercomputers.

Identifier

84990683901 (Scopus)

ISBN

[0818635312, 9780818635311]

Publication Title

Proceedings Workshop on Heterogeneous Processing Vvhp 1993 Held in Conjunction with the 7th International Parallel Processing Symposium

External Full Text Location

https://doi.org/10.1109/WHP.1993.664360

First Page

15

Last Page

22

This document is currently not available here.

Share

COinS