GMRES: a generalized minimal residual algorithm for solving nonsymmetric linear systems
SIAM Journal on Scientific and Statistical Computing
On vectorizing incomplete factorization and SSOR preconditioners
SIAM Journal on Scientific and Statistical Computing - Telecommunication Programs at U.S. Universities
Vectorization of linear recurrence relations
SIAM Journal on Scientific and Statistical Computing
CGS, a fast Lanczos-type solver for nonsymmetric linear systems
SIAM Journal on Scientific and Statistical Computing
High performance preconditioning
SIAM Journal on Scientific and Statistical Computing
(M) ICCG for 2D problems on vectorcomputers
INRIA Conference on Supercomputing: state-of-the-art
Fast iterative solution of carrier continuity equations for three-dimensional device simulation
SIAM Journal on Scientific and Statistical Computing - Special issue on iterative methods in numerical linear algebra
A Parallel Method for Tridiagonal Equations
ACM Transactions on Mathematical Software (TOMS)
LSQR: An Algorithm for Sparse Linear Equations and Sparse Least Squares
ACM Transactions on Mathematical Software (TOMS)
Three-dimensional simulation of semiconductor devices on supercomputers
ICS '91 Proceedings of the 5th international conference on Supercomputing
Hi-index | 0.00 |
Discretization and iterative solution of the semiconductor equations in a three-dimensional rectangular region lead to very large sparse linear systems. Nevertheless, design engineers and scientists of device physics need reliable results in short time in order to draw the best advantage out of computer simulation when designing new technologies and advanced devices. This goal can be achieved by use of pre-conditioned iterative methods for the solution of the linear equations on powerful computers as vector and concurrent supercomputers. To achieve optimum performance of such methods special algorithms and coding techniques have to be used in order to allow vectorization and parallelization of the inherent recurrence relations. We have investigated Jacobi and incomplete LU-factorization methods along with various hand tuning options with special emphasis on application on a SIEMENS/Fujitsu VP200. Results are compared with those obtained on a Cray-2 and an ALLIANT/FX40 minisupercomputer.