A parallel markovian model reliability algorithm for hypercube networks

https://doi.org/10.1016/0165-6074(89)90099-9Get rights and content

Abstract

Resolution of markovian models to evaluate reliability of multiprocessor systems is a operation which requires a great amount of computations when the number of nodes is high. Therefore, this kind of models are suitable for parallel procesing. In this work we present a parallel algorithm to evaluate the reliability of multiprocessor systems with degradable nodes on SIMD hypercube computers. This parallel algorithm is general in the sense that it do not impose any restriction in the problem space dimenssions (nodes, spare, degradations and time) and is adaptable to any hypercube dimension. The flexibility of the algorithm is rooted in the methodology 3evelopped by Zapata et al. [1] for embedding algorithms into hypercubes. Finally, the algorithmic complexity is analyzed.

References (12)

  • Zapata, E.L., Riviera, F.F. and Plata, O.G., On the partition of algorithms into hypercubes. In Advances on Parallel...
  • D.K. Pradhan

    Dynamically Restructurable Fault-Tolerant Processor Network Architecture

    IEEE Trans. Comput.

    (1985)
  • L. Donatello et al.

    Analysis of a Composite Performance Reliability Measure for Fault-Tolerant Systems

    Journal of the Association for Computing Machinery

    (1987)
  • A.M. Johnson et al.

    Survey of Software Tools for Evaluating Reliability, Availability and Serviceability

    ACM Computing Survey

    (1988)
  • I. Koren et al.

    Modelling the Effect of Redundancy on Yield and Performance of VLSI System

    IEEE Trans. Comput

    (1987)
  • Y.W. Ng et al.

    A Unified Reliability Model for Fault-Tolerant Computer

    IEEE Trans. Comput.

    (1980)
There are more references available in the full text version of this article.

Cited by (8)

  • Modified Gram-Schmidt QR factorization on hypercube

    1991, Journal of Parallel and Distributed Computing
  • Multidimensional fast Hartley transform onto SIMD hypercubes

    1990, Microprocessing and Microprogramming
View all citing articles on Scopus

This work was supported in part by the Minister of Education and Science (CICYT) of Spain under contract TIC88-0094

View full text