A Massively Parallel Circuit Simulator for Power Grids Analysis
Hisako Sugano, Yuuya Isoda, Makoto Yokota, Ittetsu Taniguchi, Masaya Yoshikawa, Masahiro Fukui · ITC-CSCC :International Technical Conference on Circuits Systems, Computers and Communications · 2009
With the shrinking of patterns in VLSIs, it becomes hard to neglect those problems related to IR-drop, electro migration, and so on. It takes huge calculation time to analyze and optimize these issues. The GPU computation is one of the expected approaches, since the innovation of the GPU is much faster than the CPU. The algorithm utilizes shared memories as much as possible to reduce to total computation time. The experimental results show that it achieves 15 times fast computation than CPU with similar accuracy.