Title :
Parallel one-sided Householder transformations for eigenvalues computation
Author_Institution :
Dept. de Matematica Inf., Universidade de Beira Interior, Covilha, Portugal
Abstract :
One-sided Householder transformations for reducing a matrix (symmetric or unsymmetric) to the appropriate canonical form are proposed and their advantages over the classical two-sided methods are investigated for a parallel computer with distributed memory
Keywords :
eigenvalues and eigenfunctions; mathematics computing; parallel algorithms; canonical form; distributed memory; eigenvalues computation; parallel computer; parallel one-sided Householder transformations; Computer architecture; Concurrent computing; Costs; Distributed computing; Eigenvalues and eigenfunctions; Jacobian matrices; Parallel architectures; Parallel processing; Software tools; Symmetric matrices;
Conference_Titel :
Parallel and Distributed Processing, 1993. Proceedings. Euromicro Workshop on
Conference_Location :
Gran Canaria
Print_ISBN :
0-8186-3610-6
DOI :
10.1109/EMPDP.1993.336400
