Parallel matrix computations using a reconfigurable pipelined optical bus

doi:10.1006/jpdc.1999.1569

Parallel matrix computations using a reconfigurable pipelined optical bus

1999-10-01

11

PRE

AI

DOI：10.1006/jpdc.1999.1569

Original

Original request for help

Save

Abstract

En 中文

We present fast and cost-efficient parallel algorithms for a number of important and fundamental matrix computation problems on linear arrays with reconfigurable pipelined optical bus systems. These problems include computing the inverse, the characteristic polynomial, the determinant, the rank, the Nth power, and an LU- and a QR-factorization of a matrix and solving linear systems of equations. Our algorithms provide a wide range of performance-cost combinations. Compared with known results, the running time of parallel solutions to all these problems can be reduced by a factor of O( log N) while costs are maintained under o(N-4). (C) 1999 Academic Press.

Keywords:

characteristic polynomial

cost

determinant

linear system of equations

LU-factorization

matrix inversion

matrix multiplication

optical pipelined bus

processor array. rank

time complexity

Parallel matrix computations using a reconfigurable pipelined optical bus

Abstract

Journal

Organization