Study on GPU-accelerated extraction of interconnects parasitic using CUDA and MPI

Xiaoyu Xu ,Yang Zhang Yang Zhang,Guoqiang Liu Guoqiang Liu,Wei Xu Wei Xu,Hui Qu Hui Qu

doi:10.1109/cefc.2010.5481435

Study on GPU-accelerated extraction of interconnects parasitic using CUDA and MPI

Xiaoyu Xu , Yang Zhang Yang Zhang + Show 3 more

https://doi.org/10.1109/cefc.2010.5481435

Copy DOI

Publication Date: May 1, 2010

Affiliation: Institute of Electrical Engineering, University of Technology Sydney, University of Sydney

#Graphics Processing Unit #Message Passing Interface + Show 8 more

Abstract
Full-Text PDF
Similar Papers

Abstract

Parallel computation is application-oriented, particularly for the GPU (Graphics Processing Unit) with the inherent parallelism. This paper shows the architecture of a GPU cluster based on MPI (Message Passing Interface) and CUDA (Compute Unified Device Architecture). Results show that the acceleration ratio is obviously improved but the acceleration effect seems decelerated in large-scale GPU cluster. The parallel algorithm is mainly focused on task partitioning sparse matrix-vector multiplications (SpVM) in GPUs.

Full Text