A new stabilization strategy for incomplete LU preconditioning of indefinite matrices

Authors:
Li Wang;Jun Zhang
Affiliations:
Department of Computer Science, Laboratory for High Performance Scientific Computing and Computer Simulation, University of Kentucky, 773 Anderson Hall, Lexington, KY;Department of Computer Science, Laboratory for High Performance Scientific Computing and Computer Simulation, University of Kentucky, 773 Anderson Hall, Lexington, KY
Venue:
Applied Mathematics and Computation
Year:
2003

Citing 3
Cited 2

Experimental study of ILU preconditioners for indefinite matrices

Journal of Computational and Applied Mathematics
A Multilevel Dual Reordering Strategy for Robust Incomplete LU Factorization of Indefinite Matrices

SIAM Journal on Matrix Analysis and Applications
Iterative Methods for Sparse Linear Systems

Iterative Methods for Sparse Linear Systems

Stabilized approximate inverse preconditioners for indefinite matrices

Proceedings of the 46th Annual Southeast Regional Conference on XX
A two-phase preconditioning strategy of sparse approximate inverse for indefinite matrices

Computers & Mathematics with Applications

Quantified Score

Hi-index	0.49

Visualization

Abstract

A two step combined preconditioning strategy is proposed to construct stable and accurate incomplete LU factorization of indefinite matrices arising from CFD applications. This preconditioning procedure is divided into two steps, each step is a factorization of a shifted matrix. Numeral experiments show that a preconditioner can be computed with high accuracy and low fill-in, and that the new strategy is robust on some difficult indefinite matrix test problems.