The block-bordered class of nonlinear systems is typically related to the modeling of medium-to-large size problems, to which domain decomposition techniques are applied. Very frequently this is a preliminary phase of an iterative approach to the solution. The PICRN algorithm by Feng and Schnabel is an implicit, two-stage quasi-Newton method that takes advantage from the structure of the Jacobian system and is very suitable for distributed memory multiprocessor architectures. A Cray T3D experimentation carried out on a nonlinear, particularly ill-conditioned system, coming from a VLSI simulation, allows to study the algorithm behaviour and gives the reason for the needed improvements and theoretical extensions successively proposed.
Cray T3D numerical experience on block bordered nonlinear systems
ZANGHIRATI, Gaetano;RUGGIERO, Valeria
1998
Abstract
The block-bordered class of nonlinear systems is typically related to the modeling of medium-to-large size problems, to which domain decomposition techniques are applied. Very frequently this is a preliminary phase of an iterative approach to the solution. The PICRN algorithm by Feng and Schnabel is an implicit, two-stage quasi-Newton method that takes advantage from the structure of the Jacobian system and is very suitable for distributed memory multiprocessor architectures. A Cray T3D experimentation carried out on a nonlinear, particularly ill-conditioned system, coming from a VLSI simulation, allows to study the algorithm behaviour and gives the reason for the needed improvements and theoretical extensions successively proposed.I documenti in SFERA sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.