The University of Southampton
University of Southampton Institutional Repository

A general purpose parallel computer

A general purpose parallel computer
A general purpose parallel computer

The processor-array is a parallel computer consisting of an interconnected array of processors sharing a single controller. The controller is similar to that of a conventional computer, whilst the processors are usually kept simple. Consequently the processor-array is relatively easy to implement, making it an attractive parallel architecture. However, there are disadvantages to this approach which have tended to restrict the areas of application of the processor-array to those requiring only simple operations on very large arrays of data. This thesis describes a processor-array architecture with novel features that overcome these limitations. This architecture is called the Reconfigurable Processor-Array (RPA). It has the same advantages as a conventional processor array - a single controller and many identical, simple, processors - but it is a more general purpose architecture. The RPA can process arrays of data smaller than the array of processors, a flexibility achieved by allowing clusters of processors to operate on each element of data. The processors themselves have been improved to make them suitable for complex operations by the addition of specialised hardware. Particular emphasis has been placed on floating-point arithmetic, which is notoriously inefficient on the simple processors of processor-arrays. The implementation of a chip containing 16 identical processors for the RPA is described. An array can be built using just this processor chip and conventional RAM chips. (D82123)

University of Southampton
Rushton, Andrew John
Rushton, Andrew John

Rushton, Andrew John (1987) A general purpose parallel computer. University of Southampton, Doctoral Thesis.

Record type: Thesis (Doctoral)

Abstract

The processor-array is a parallel computer consisting of an interconnected array of processors sharing a single controller. The controller is similar to that of a conventional computer, whilst the processors are usually kept simple. Consequently the processor-array is relatively easy to implement, making it an attractive parallel architecture. However, there are disadvantages to this approach which have tended to restrict the areas of application of the processor-array to those requiring only simple operations on very large arrays of data. This thesis describes a processor-array architecture with novel features that overcome these limitations. This architecture is called the Reconfigurable Processor-Array (RPA). It has the same advantages as a conventional processor array - a single controller and many identical, simple, processors - but it is a more general purpose architecture. The RPA can process arrays of data smaller than the array of processors, a flexibility achieved by allowing clusters of processors to operate on each element of data. The processors themselves have been improved to make them suitable for complex operations by the addition of specialised hardware. Particular emphasis has been placed on floating-point arithmetic, which is notoriously inefficient on the simple processors of processor-arrays. The implementation of a chip containing 16 identical processors for the RPA is described. An array can be built using just this processor chip and conventional RAM chips. (D82123)

This record has no associated files available for download.

More information

Published date: 1987

Identifiers

Local EPrints ID: 461831
URI: http://eprints.soton.ac.uk/id/eprint/461831
PURE UUID: 99e99813-c1ba-4311-89c1-af2e839cc6c3

Catalogue record

Date deposited: 04 Jul 2022 18:56
Last modified: 04 Jul 2022 18:56

Export record

Contributors

Author: Andrew John Rushton

Download statistics

Downloads from ePrints over the past year. Other digital versions may also be available to download e.g. from the publisher's website.

View more statistics

Atom RSS 1.0 RSS 2.0

Contact ePrints Soton: eprints@soton.ac.uk

ePrints Soton supports OAI 2.0 with a base URL of http://eprints.soton.ac.uk/cgi/oai2

This repository has been built using EPrints software, developed at the University of Southampton, but available to everyone to use.

We use cookies to ensure that we give you the best experience on our website. If you continue without changing your settings, we will assume that you are happy to receive cookies on the University of Southampton website.

×