An Improved Parallel Implementation of 3D DRIE Simulation on GPU
Nankai-Baidu joint lab, College of Information Technical Science, Nankai University, Weijin Road 94, Tianjin, 300071, China
10th International Symposium on Pervasive Systems, Algorithms, and Networks (ISPAN), 2009
@conference{zhang2009improved,
title={An Improved Parallel Implementation of 3D DRIE Simulation on GPU},
author={Zhang, F. and Wang, G. and Liu, X. and Liu, J.},
booktitle={2009 10th International Symposium on Pervasive Systems, Algorithms, and Networks},
pages={747–751},
year={2009},
organization={IEEE}
}
Deep reactive ion etching (DRIE) technique is a new and powerful tool in Micro-Electro-Mechanical Systems (MEMS) fabrication. A 3D DRIE simulation can help researcher understand the time-evolution of Bosch process used in DRIE. Due to the high complexity of the algorithm used in the simulation, it is necessary to develop an algorithm that can accelerate the simulation. This paper presents a parallel implementation of the 3D DRIE simulation based on GPU, built on Nvidia’s Compute Unified Device Architecture (CUDA) platform. This paper also presents a fast morphological operation, which reduces the complexity of mathematical morphology operation part of the algorithm from O(N3) to O(N2). The experiment results show the parallel program on Nvidia GTX260+ GPU obtains about 70x to 75x speedup over the 4-threads parallel version on Intel Q6600 CPU.
May 7, 2011 by hgpu
