|
The innermost computational
kernel of many large industrial numerical simulations is a large sparse matrix linear
system. A recent survey indicated that more than 70% of supercomputer time is used
for solving large linear systems. A great impact can be made if performance of these
sparse matrix solvers can be improved.
Storage requirements for three-dimensional simulations make direct methods prohibitively
expensive. Iterative techniques become mandatory but lack the robustness of direct
methods. Past emphasis has been devoted to exploring more powerful iterative solvers.
These techniques did not have much initial impact, and experience indicates that
a good preconditioner holds the key to an effective iterative solver.
This conference, sponsored by the University of Minnesota Supercomputing Institute
and the University of Waterloo in cooperation with SIAM Linear Algebra Group, brings
researchers and engineers together to discuss recent developments and progress made
as well as to exchange findings and explore possible new directions in this field.
Invited speakers include Raymond Chan from The Chinese University of Hong Kong, Edmond
Chow from Lawrence Livermore National Laboratory, Howard Elman from the University
of Maryland, Charbel Farhat from the University of Colorado, Peter Forsyth from the
University of Waterloo in Canada, David Keyes from the NASA Langley Research Center,
John Lewis and Dan Pierce from Boeing Computer Service, Maya Neytcheva from the University
of Nijimegen in The Netherlands, Willy Schilders from Philips Research Laboratories,
Henk van der Vorst from Utrecht University in the Netherlands, and Justin Wan from
Stanford University. Contributed papers are being solicited.
Program Committee
Yousef Saad, University of Minnesota, Conference Chair
Daniel Pierce, Boeing Company, Program Co-Chair
Wei-Pai Tang, University of Waterloo, Program Co-Chair
Steve Ashby, Lawrence Livermore National Laboratory
Alda Behie, Intera Consultants Limited
Ake Bjorck, Linköping University, Sweden
Tony Chan, University of California at Los Angeles
Iain Duff, Rutherford Appleton Laboratory, United Kingdom
John Gilbert, Xerox Palo Alto Research Center
Esmond Ng, Oak Ridge National Laboratory
Barry Smith, Argonne National Laboratory
Horst Simon, Lawrence Berkeley National Laboratory
Henk van der Vorst, Utrecht University, The Netherlands
Registration and Contributed Papers
Information about registration and contributing papers is available by sending
email to sparse99@msi.umn.edu or at:
www2.msi.umn.edu/Symposia/sparse99 |
|
|