Positive definiteness aspects of vectorizable preconditioners

DOI10.1016/0167-8191(89)90080-XMaRDI QIDQ583846zbMATH OpenFDO

Authors Victor Eijkhout, Panayot S. Vassilevski

Publication date 1989

Published in Parallel Computing (Search for Journal in Brave)

Full work available at URL https://doi.org/10.1016/0167-8191(89)90080-X

zbMATH Keywords

conjugate gradient iterationincomplete block factorization methodsM-matricespositive definitenesspreconditionerrow sum criteriatest problemsvectorization

Mathematics Subject Classification ID

Boundary value problems for second-order elliptic equations (35J25) Iterative numerical methods for linear systems (65F10) Numerical computation of matrix norms, conditioning, scaling (65F35) Numerical solution of discretized equations for boundary value problems involving PDEs (65N22)

Review text

Two main classes of incomplete block factorization methods for solving discrete elliptic problems are considered. One of them is more suitable for vectorization, but loss of positive definiteness of the preconditioner, and failure of the conjugate gradient iteration might occur even for M-matrices. In order to improve the chances that this will not occur, certain row sum criteria are often used. It is shown in this paper, both theoretically and experimentally, that this does not solve the underlying problem. In fact, for modestly difficult test problems, the methods considered do not perform very satisfactorily at all.

Recommendations

• Uniform preconditioners for problems of positive order
• The use of preconditioning in direct methods of solving positive definite linear systems
• Multisplitting preconditioners for a symmetric positive definite matrix
• A robust incomplete factorization preconditioner for positive definite matrices
• The preconditioned SOR iterative method for positive definite matrices
• scientific article; zbMATH DE number 4190984
• Massively parallel preconditioners for symmetric positive definite linear systems
• A symmetric positive definite preconditioner for saddle-point problems
• Combination preconditioning of saddle point systems for positive definiteness.
• scientific article; zbMATH DE number 877150

Cited in

(6)

• Vectorizable preconditioners for elliptic difference equations in three space dimensions
• A vectorizable variant of pgcr methods for unsymmetric linear systems
• Vectorizable preconditioners for mixed finite element solution of second-order elliptic problems
• Incomplete block-factorization preconditioners for solving three-dimensional elliptic difference equations on systolic processors^∗∗∗
• The pgcr method for solving unsymmetric linear systems on a vector multiprocessor
• On some ways of approximating inverses of banded matrices in connection with deriving preconditioners based on incomplete block factorizations

This page was built for publication: Positive definiteness aspects of vectorizable preconditioners

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q583846)