Volume 52, Issue 3
Partial Fraction Decomposition of Matrices and Parallel Computing

Fredéric Hecht & Sidi-Mahmoud Kaber

J. Math. Study, 52 (2019), pp. 244-257.

Published online: 2019-09

Preview Full PDF 55 2622
Export citation
  • Abstract

We are interested in the design of parallel numerical schemes for linear systems. We give an effective solution to this problem in the following case: the matrix $A$ of the linear system is the product of $p$ nonsingular matrices $A_i^m$ with specific shape: $A_i= I -{h_i}X$ for a fixed matrix $X$ and real numbers $h_i$. Although having a  special form, these matrices $A_i$ arise frequently in the discretization of evolutionary Partial Differential Equations. For example, one step of the implicit Euler scheme for the evolution equation $u' =Xu$ reads $(I-hX)u^{n+1}=u^n$. Iterating $m$ times such a scheme leads to a linear system $Au^{n+m}=u^n$. The idea is to express $A^{-1}$ as a linear combination of elementary  matrices $A_i^{-1}$ (or more generally in term of matrices $A_i^{-k}$). Hence the solution of the linear system with matrix $A$ is a linear combination of the solutions of  linear systems with matrices $A_i$ (or $A_i^k$). These systems are then solved simultaneously on different processors.

  • Keywords

Partial differential equations, parabolic equation, finite element methods, finite difference methods, parallel computing.

  • AMS Subject Headings

65M60, 65Y05, 35K45, 74S05, 74S20

  • Copyright

COPYRIGHT: © Global Science Press

  • Email address

hecht@ljll.math.upmc.fr (Fredéric Hecht)

kaber@ljll.math.upmc.fr (Sidi-Mahmoud Kaber)

  • BibTex
  • RIS
  • TXT
Copy to clipboard
The citation has been copied to your clipboard