The domain decomposition paradigms of Schwarz and Schur are considered, along with various hybrids, and their superiority in terms of data motion is noted relative to some standard decompositions. A number of highlights from computational mechanics codes on the ASCI platforms are described, along with freely available software packages implementing some of them. Research issues motivated by the need for looser coupling and fault tolerance are thrown out for consideration.
David Keyes is the Richard F. Barry Professor of Mathematics and Statistics at Old Dominion University, an Adjunct Professor of Computer Science at ODU, the Director of the Center for Computational Sciences at ODU, the Acting Director of Institute for Scientific Computing Research (ISCR) at the Lawrence Livermore National Laboratory, and an Associate Research Fellow of ICASE at the NASA Langley Research Center. Keyes received his Ph.D. in Applied Mathematics from Harvard University in 1984. He has an M.S. in Applied Mathematics, also from Harvard, awarded in 1979, and he graduated summa cum laude with a B.S.E. in Aerospace and Mechanical Sciences and a Certificate in Engineering Physics from Princeton University in 1978.
Keyes is the author or co-author of more than 90 publications in computational science, numerical analysis, and computer science. He has co-edited 7 conference proceedings concerned with parallel algorithms. Keyes has delivered over 200 presentations at universities, laboratories, and industrial research centers in 22 countries. With backgrounds in engineering, applied mathematics, and computer science, and consulting experience with industry and national laboratories, Keyes works at the algorithmic interface between parallel computing and the numerical analysis of partial differential equations, across a spectrum of aerodynamic, geophysical, and chemically reacting flows. Newton-Krylov-Schwarz parallel implicit methods, introduced in a 1993 paper he co-authored at ICASE, are now widely used throughout engineering and computational physics, and have been scaled to thousands of processors on the ASCI platforms.
Keyes has co-organized or lectured in numerous conferences and short courses on high-performance computing for systems modeled by PDEs for NASA Langley, LLNL, SIAM, the DoD Modernization Centers, the domain decomposition and parallel CFD communities, and university departments. He is currently a member of the editorial boards of Parallel and Distributed Computing Practices, Int. J. High Performance Computing Applications, and Springer's Lecture Notes in Computational Science and Engineering and has served as an editor of SIAM J. Scientific Computing.
Among Keyes' awards are: the Gordon Bell Prize for High Performance Computing, Special Category (shared), 1999; the National Science Foundation Presidential Young Investigator Award, 1989; the Yale College Prize for Teaching Excellence in the Natural Sciences, 1991; a Yale University Junior Faculty Fellowship, 1990 - 91; and a Harvard-Danforth Certificate for Excellence in Teaching, 1982. He has led one of the 37 NSF "Grand, National, and Computational Challenge" Centers and one of the 14 DOE ASCI "Level 2" Centers. He currently directs a nine-institution Integrated Software Infrastructure Center (ISIC) for the Office of Advanced Scientific Computing Research of the DOE, one of seven such centers nationally under the Scientific Discovery through Advanced Computing (SciDAC) initiative.