Iterative Solution of Large Sparse Systems of Equations / Edition 1

by Wolfgang Hackbusch

Overview

This book presents the description of the state of the modern iterative techniques together with systematic analysis. It discusses classical methods, semi-iterative techniques, incomplete decompositions, conjugate gradient methods, multigrid methods and domain decomposition techniques. Every technique described is illustrated by a Pascal program.

Product Details

ISBN-13:	9781461287247
Publisher:	Springer New York
Publication date:	09/27/2011
Series:	Applied Mathematical Sciences , #95
Edition description:	Softcover reprint of the original 1st ed. 1994
Pages:	432
Product dimensions:	6.10(w) x 9.25(h) x 0.04(d)

About the Author

Wolfgang Hackbusch is a Professor in the Scientific Computing department at Max Planck Institute for Mathematics in the Sciences. His research areas include numerical treatment of partial differential equations, numerical treatment of integral equations, and hierarchical matrices.

Table of Contents

Part I: Linear Iterations.- Introduction.- Iterative Methods.- Classical Linear Iterations in the Positive Definite Case.- Analysis of Classical Iterations Under Special Structural Conditions.- Algebra of Linear Iterations.- Analysis of Positive Definite Iterations.- Generation of Iterations. Part II: Semi-Iterations and Krylov Methods.- Semi-Iterative Methods.- Gradient Methods.- Conjugate Gradient Methods and Generalizations.- Part III: Special Iterations.- Multigrid Iterations.- Domain Decomposition and Subspace Methods.- H-LU Iteration.- Tensor-based Methods.- Appendices.

Editorial Reviews

Translated from the German edition. The author describes the recent state of iterative and related methods without delving into specialized areas. Pascal programs are explicitly given and a collection of the source codes are available on disk from the author. Topics include analysis in the 2-cycle case, analysis for m- matrices, semi-iterative methods, transformations, secondary iterations, incomplete triangular decompositions, conjugate gradient methods, multi-grid methods, and domain decomposition methods. Exercises are included. Annotation c. Book News, Inc., Portland, OR (booknews.com)

Booknews