Proceedings ArticleDOI
Massively parallel device simulation using irregular grids
Jagesh V. Sanghavi,E. Tomacruz,Alberto Sangiovanni-Vincentelli +2 more
- pp 141-144
Reads0
Chats0
TLDR
Partitioning, communication scheduling, and preconditioning algorithms are developed to implement a parallel linear solver for an irregular grid drift-diffusion device simulator and are compared with PILS running on a single processor.Abstract:
Partitioning, communication scheduling, and preconditioning algorithms are developed to implement a parallel linear solver for an irregular grid drift-diffusion device simulator. The parallel program is executed on a 64 node CM-5 and is compared with PILS running on a single processor. We obtain an average CPU time speed-up of 46.1X for each CGS iteration with no preconditioning, and a speed-up of 33.6X for the solution of the matrix. >read more
References
More filters
Journal ArticleDOI
Partitioning of unstructured problems for parallel processing
TL;DR: Numerical comparisons on large-scale two- and three-dimensional problems demonstrate the superiority of the new spectral bisection algorithm.
Journal ArticleDOI
Three-dimensional numerical semiconductor device simulation: algorithms, architectures, results
TL;DR: The authors present SECOND, a program for large-scale semiconductor device simulation with truly three-dimensional grids, which investigated the most commonly used numerical algorithms for the solution of the classical drift-diffusion equations.
Proceedings ArticleDOI
Connection Machine model CM-5 system overview
J. Palmer,G.L. Steele +1 more
TL;DR: The Connection Machine model CM-5 provides high performance and ease of use for large data-intensive applications and is designed to scale to teraflops performance on terabyte-sized problems.
Proceedings ArticleDOI
PILS: an iterative linear solver package for ill-conditioned systems
C. Pommerell,W. Fichtner +1 more
TL;DR: In this article, the authors propose a method to solve the problem of unstructured data in the context of data augmentation, and propose an approach based on the concept of self-healing.
Journal ArticleDOI
A STRIDE towards practical 3-D device simulation-numerical and visualization considerations
TL;DR: A 3-D device solver, capable of solving grids up to 250,000 nodes, has been developed on a message passing multiprocessor and by using an independent-edge-grouping scheme to increase the vector length to the order of the number of variables, the vector processing efficiency is significantly increased without additional floating point operations.