Artificial Neural Network Simulation on CUDA
CloudCore Research Group, School of Computing, Dublin City University, Dublin, Ireland
The 16th IEEE/ACM International Symposium onDistributed Simulation and Real Time Applications (DS-RT ’12), 2012
@inproceedings{pendlebury2012artificial,
title={Artificial Neural Network Simulation on CUDA},
author={Pendlebury, John and Xiong, Huanhuan and Walshe, Ray},
booktitle={Proceedings of the 2012 IEEE/ACM 16th International Symposium on Distributed Simulation and Real Time Applications},
pages={228–233},
year={2012},
organization={IEEE Computer Society}
}
The advent of low cost GPU hardware and user friendly parallel programming APIs, such as NVIDIA CUDA means that affordable, programmable, high-performance computing environments for simulation are now attainable for development of scientific simulations. In this paper the authors present the MineHunter program, a parallel simulation of neural networks on NVIDIA CUDA. The simulation consists of 128 mine hunters in a mine field of 8192 mines, running on an Intel QuadCore i5-2500 3.3GHz 2 x Nvidia GeForce GTX 480. The results presented demonstrate that CUDA improves performance by up to 80% compared with the equivalent CPU implementation.
June 21, 2013 by hgpu
