CAREER: Robust Preconditioners for Sparse Linear Systems

Texas A&M Engineering Experiment Station, College Station TX

Investigators

Abstract

This project will develop a new class of robust and parallelizable preconditioners for the solution of sparse linear systems arising from partial differential equations. The approach is based on a hierarchical decomposition of the matrices associated with the linear system, and is designed to improve the effectiveness of the preconditioner without compromising the parallelism inherent in the computation. A formal methodology will be developed for construction, application, and analysis of these preconditioners. Effectiveness of these techniques will be tested on problems in incompressible fluid simulations. Parallel implementations of these preconditioners will be developed and their performance evaluated for various multiprocessor platforms. Parallel libraries will also be developed and made available to others via the internet. As part of the educational plan innovative techniques will be employed for curriculum development in computational science and engineering, and university wide workshops and seminars will be organized for the research community.