The authors present SECOND, a program for large-scale semiconductor device simulation with truly three-dimensional grids. Since 3-D simulations necessitate large computing resources, the choice of algorithms and their implementation become of utmost importance. The authors investigated the most commonly used numerical algorithms for the solution of the classical drift-diffusion equations. The study included coupled and noncoupled point and block schemes, direct and preconditioned iterative linear solvers, and several distinct ordering and coloring techniques. Structures with regular and irregular grids were analyzed. These algorithms were compared on a variety of machines including workstations, minisupers, and supercomputers. Results of transient simulations are presented to illustrate the approach.>

We have simulated several numerical and non-numerical algorithms on five distributed-memory parallel processors (DMPPs). All five DMPPs have the same topology (a torus), and the same number of nodes. The architectures differ only in the communication speed between neighboring nodes, while the computation unit is kept unchanged. The goal of the paper is to quantify the effect that interprocessor communication speed and synchronization overhead have on the performance of the DMPPs. After introducing the rationale for this study and reviewing related work, we present and discuss the results of the simulations.

K9 is a software package for the simulation and performance evaluation of distributed-memory parallel processors (DMPPs). It is written in C++ and runs on Sequent Symmetry and SUN-3. K9 provides the user with four building-blocks (processor cells, communication channels, multi-port shared-memories, and I/O processors), and one abstraction mechanism (the DMPP interconnection topology). Application code for K9 can be written in C++ or C.