Fault-Tolerant Multiprocessor and VLSI-Based Systems.

Abstract

Study and development of certain fault-tolerant architectures that utilize the capabilities of new IC technology was undertaken. Specifically, the research was aimed at network architectures, distinguished by a close interconnection of a large number of computing elements. Included is a subclass of specialized network architectures known as VLSI processor arrays. Besides fault-tolerance-related research for such arrays, also proposed was exploration of a new array architecture, developed for the express purpose of executing general algorithms on these arrays. The precise research formulated-developing fault-tolerant multiprocessor network architectures-goes beyond earlier work. Here, the system interconnection structure, itself, was used as the primary design tool for achieving various and diverse objectives, including: low interconnection and layout complexities, dynamic reconfigurability, fault-tolerance through graceful degradation as well as self-diagnosability. Viability of the proposed research was demonstrated in the proposal; new communication structures were introduced, along with concepts of admissability of multiple logical configurations, and algorithmic and detour routing that provide fault-tolerance and graceful degradation.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Mar 15, 1987
Accession Number
ADA183820

Entities

People

  • Dhiraj K. Pradhan

Organizations

  • University of Massachusetts Amherst

Tags

Communities of Interest

  • Energy and Power Technologies

DTIC Thesaurus Topics

  • Algorithms
  • Arrays
  • Bus Networks
  • Coding
  • Computational Complexity
  • Computer Networks
  • Computers
  • Computing System Architectures
  • Engineering
  • Failure Mode And Effect Analysis
  • Fault Tolerance
  • Fault Tolerant Computing
  • Linear Arrays
  • Network Architecture
  • Parallel Computing
  • Parallel Processing
  • Trees (Data Structures)

Fields of Study

  • Engineering

Readers

  • Parallel and Distributed Computing.
  • Systems Analysis and Design