Advanced Parallel Systems

Abstract

The objective of this work was to explore algorithms and their implementation for future advanced parallel systems. These systems are assumed to have hundreds or even thousands of processors and to be able to concentrate their computing power on one or a small number of tasks. The three principal questions to be explored were: 1) Are there algorithms for the crucial applications which have enough parallelism to allow the power of the advanced parallel systems to be fully exploited?; 2) What languages and implementation tools are needed for efficient programming of these algorithms?; 3) What are the relative performances of different algorithm types? Of different architecture types? Of different implementation languages? The research results obtained appear in this report.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Nov 01, 1989
Accession Number
ADA217175

Entities

People

  • John Rice

Organizations

  • Purdue University

Tags

Communities of Interest

  • Energy and Power Technologies

DTIC Thesaurus Topics

  • Algorithms
  • Boundary Value Problems
  • Computational Science
  • Computer Programming
  • Computer Programs
  • Computer Science
  • Computers
  • Differential Equations
  • Equations
  • Language
  • Parallel Computing
  • Parallel Processing
  • Parallel Processors
  • Partial Differential Equations
  • Petri Nets
  • Software Development

Fields of Study

  • Computer science

Readers

  • Computational Modeling and Simulation
  • Computer Science.
  • Neural Network Machine Learning.