Share Email Print

Proceedings Paper

Algorithm-based Fault Tolerance for Parallel Matrix Equation Solvers
Author(s): Franklin T. Luk
Format Member Price Non-Member Price
PDF $14.40 $18.00

Paper Abstract

We examine the checksum schemes of Abraham et al. for the computation of the LU-factorization using a multiprocessor array. Their methods are very efficient for detecting a transient error, but quite expensive for correcting it due to the need for a computation rollback. In this paper, we show how to avoid the rollback and how to implement pivoting. We also introduce a new checksum method for solving triangular sets of linear equations.

Paper Details

Date Published: 4 January 1986
PDF: 7 pages
Proc. SPIE 0564, Real-Time Signal Processing VIII, (4 January 1986); doi: 10.1117/12.949703
Show Author Affiliations
Franklin T. Luk, Cornell University (United States)

Published in SPIE Proceedings Vol. 0564:
Real-Time Signal Processing VIII
Keith Bromley; William J. Miceli, Editor(s)

© SPIE. Terms of Use
Back to Top