New evaluation index of incomplete Cholesky preconditioning effect

Authors:
Takeshi Iwashita;Masaaki Shimasaki
Affiliations:
Academic Center for Computing and Media Studies, Kyoto University, Kyoto, Japan;Graduate School of Engineering, Kyoto University, Kyoto, Japan
Venue:
ISHPC'05/ALPS'06 Proceedings of the 6th international symposium on high-performance computing and 1st international conference on Advanced low power systems
Year:
2005

Citing 6
Cited 0

The effect of ordering on preconditioned conjugate gradients

BIT
Applied numerical linear algebra

Applied numerical linear algebra
Ordering strategies and related techniques to overcome the trade-off between parallelism and convergence in incomplete factorizations

Parallel Computing - Special Anniversary issue
Three-Dimensional Finite Brick-Type Edge-Element Eddy Current Analysis Using Parallelized Linear-System Solvers

PARELEC '00 Proceedings of the International Conference on Parallel Computing in Electrical Engineering
Iterative Methods for Sparse Linear Systems

Iterative Methods for Sparse Linear Systems
Comparison Criteria for Parallel Orderings in ILU Preconditioning

SIAM Journal on Scientific Computing

Quantified Score

Hi-index	0.00

Visualization

Abstract

In Incomplete LU (ILU) preconditioning, orderings often affect the effect of preconditioning. The authors recently proposed a simple evaluation way for orderings in the ILU preconditioning technique. The present paper introduces the evaluation method in unstructured analyses in which the effect of preconditioning is not easily estimated. The evaluation index, which has a simple relationship with the matrix norm of the remainder matrix, is easily computed without additional memory requirement. The computational cost of the index is trivial in the total iterative solution process. The effectiveness of the method is examined by numerical tests using coefficient matrix data from the Matrix Market, a finite-difference analysis of Poisson equation, and a 3-d electromagnetic field analysis.