Title :
Matrix sign function implementations on a hypercube multiprocessor
Author :
Gardiner, Judith D. ; Laub, A.
Author_Institution :
Dept. of Electr. & Comput. Eng., California Univ., Santa Barbara, CA, USA
Abstract :
Three implementations of a matrix sign function algorithm for solving generalized algebraic Riccati equations on an Intel iPSC/d5 hypercube multiprocessor are described. The three implementations are compared for complexity of computation and communication. The best one showed efficiencies of about 60% on large problems. Timing results are given
Keywords :
mathematics computing; matrix algebra; multiprocessing systems; Intel iPSC/d5 hypercube multiprocessor; generalized algebraic Riccati equations; matrix sign function algorithm; Algorithm design and analysis; Concurrent computing; Hypercubes; Iterative algorithms; Parallel algorithms; Parallel processing; Partitioning algorithms; Riccati equations; Symmetric matrices; Timing;
Conference_Titel :
Decision and Control, 1988., Proceedings of the 27th IEEE Conference on
Conference_Location :
Austin, TX
DOI :
10.1109/CDC.1988.194569
