A Lazy, Self-optimising Parallel Matrix Library

Abstract

This paper describes a parallel implementation of a matrix/vector library for C++ for a large distributed-memory multicomputer. The library is “self-optimising” by exploiting lazy evaluation: execution of matrix operations is delayed as much as possible. This exposes the context in which each intermediate result is used. The run-time system extracts a functional representation of the values being computed and optimises data distribution, grain size and scheduling prior to execution. This exploits results in the theory of program transformation for optimising parallel functional programs, while presenting an entirely conventional interface to the programmer. We present details of some of the simple optimisations we have implemented so far and illustrate their effect using a small example.

Contributors

Paul H J Kelly

Conference

Publication date: July 1995

Publication date (Print): July 1995

Pages: 1-10

Affiliations

Department of Computing, Imperial College

180 Queen’s Gate, London SW7 2BZ, UK

Article

DOI: 10.14236/ewic/FP1995.8

Copyright statement: © Simon Govier et al. Published by BCS Learning and Development Ltd. Proceedings of the 1995 Glasgow Workshop on Functional Programming, Ullapool, Scotland

License:

This work is licensed under a Creative Commons Attribution 4.0 Unported License. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/

Conference name: Proceedings of the 1995 Glasgow Workshop on Functional Programming

Conference acronym: FP

Conference number:

Conference location: Ullapool, Scotland

Conference date: 10-12 July 1995

Conference sponsor: Electronic Workshops in Computing (eWiC)

Conference theme: Functional Programming

Product

Product Information: 1477-9358BCS Learning & Development

Self URI (article page): https://www.scienceopen.com/hosted-document?doi=10.14236/ewic/FP1995.8

Self URI (journal page): https://ewic.bcs.org/

Categories

Series title: Electronic Workshops in Computing