2006Research Repository (Delft University of Technology)Open access

Multilevel Iterative Methods and Deflation

Radim Blaheta

Open full text 2 citations

Abstract

In this paper, we show that the deflation method can be viewed as a possible implementation of the CG method with multilevel preconditioner. Further, we demonstrate efficiency and robustness of different implementations of multilevel preconditioners with different "coarse grid" spaces by solving a simple model problem.

Open-access reader

About this research paper

What this paper is about

In this paper, we show that the deflation method can be viewed as a possible implementation of the CG method with multilevel preconditioner. Further, we demonstrate efficiency and robustness of different implementations of multilevel preconditioners with different "coarse grid" spaces by solving a simple model problem.

Why it matters

OpenAlex reports 2 citations for this work. Citation counts describe recorded attention and do not establish research quality.

Key contribution

A contribution statement is not available in the OpenAlex record.

Method / approach

Method details are not available in the OpenAlex metadata.

Main findings

Findings are not separately available in the OpenAlex metadata.

Limitations

Limitations are not available in the OpenAlex metadata.

Applications

Application details are not available in the OpenAlex metadata.

Available abstract

In this paper, we show that the deflation method can be viewed as a possible implementation of the CG method with multilevel preconditioner. Further, we demonstrate efficiency and robustness of different implementations of multilevel preconditioners with different "coarse grid" spaces by solving a simple model problem.

Key concepts: Preconditioner, Deflation, Robustness (evolution), Computer science, Implementation, Grid, Simple (philosophy), Iterative method

Related papers

Back to paper searchBrowse research topicsOriginal source
Multilevel Iterative Methods and Deflation — Research Paper | ScholarLens