Toggle Main Menu Toggle Search

Open Access padlockePrints

Fail-controlled Computer Architectures for Distributed Systems

Lookup NU author(s): Emeritus Professor Santosh Shrivastava, Dr Paul EzhilchelvanORCiD, Dr Neil Speirs



The problem of constructing reliable processors, hereafter referred to as nodes, out of conventional unreliable processors is examined. We assume that a node is composed of a number of processors and failed processors can exhibit arbitrary behavious. The non-faulty processors o a node execute Byzntine agreement protocols to maintain the abstraction of a fail-controlled (well-behaved) node. the paper presents a variety of node architectures for maintaining the abstraction of fail controlled behaviour with differing functionalities. Some of these architectures are shown to be of practical interest.

Publication metadata

Author(s): Shrivastava SK, Ezhilchelvan PD, Speirs NA, Seaton DT

Publication type: Report

Publication status: Published

Series Title: Computing Laboratory Technical Report Series

Year: 1991

Pages: 18

Print publication date: 01/07/1991

Source Publication Date: July 1991

Report Number: 333

Institution: Computing Laboratory, University of Newcastle upon Tyne

Place Published: Newcastle upon Tyne