Research on parallel methods, algorithms, and software at the University of Texas at Austin

This is an overview of some of the research concerned with parallel methods, algorithms, and software carried out in the Center for Numerical Analysis and Computational Fluid Dynamics Laboratory of The University of Texas at Austin. Of particular interest is the solution of very large scale problems involving partial differential equations where finite element and finite different methods with iterative solvers are used on high-performance supercomputers. Such applications arise, for instance, in computational fluid dynamics and lead to nonsymmetric sparse algebraic systems. Part of the research involves the development of new and improved methods and ways of mapping them onto parallel architectures as well as the implementation of these schemes in software packages and tests involving representative applications. Since we focus on work related to our research centers, the treatment is not exhaustive nor is the bibliography.