Submitted Presentations

PARLE '89 Parallel Architectures and Languages Europe

Volume 365 of the series Lecture Notes in Computer Science pp 193-206

Date:

High-performance parallel graph reduction

  • Simon L Peyton JonesAffiliated withDepartment of Computer Science, University College London
  • , Chris ClackAffiliated withDepartment of Computer Science, University College London
  • , Jon SalkildAffiliated withDepartment of Computer Science, University College London

* Final gross prices may vary according to local VAT.

Get Access

Abstract

Parallel graph reduction is an attractive implementation for functional programming languages because of its simplicity and inherently distributed nature. This paper outlines some of the issues raised by parallel compiled graph reduction, and presents the approach we have adopted for our parallel machine, GRIP.

We concentrate on two main areas:
  • •Static and dynamic techniques to control the growth of parallelism, so as to provide enough parallelism of an appropriate granularity to keep the machine busy without swamping it.

  • •Dynamic techniques to exploit the memory hierarchy, so that frequently-referenced data is held near to the processor that references it.