On the Effect of Using Multiple GPUs in Solving QAPs with CUDA
Hannan University, 5-4-33, Amami-higashi, Matsubara, Osaka 580-8502, Japan
Fourteenth international conference on Genetic and evolutionary computation conference companion (GECCO Companion ’12), 2012
@inproceedings{tsutsui2012effect,
title={On the effect of using multiple GPUs in solving QAPs with CUDA},
author={Tsutsui, Shigeyoshi and Fujimoto, Noriyuki},
booktitle={Proceedings of the fourteenth international conference on Genetic and evolutionary computation conference companion},
pages={629–630},
year={2012},
organization={ACM}
}
In this paper, we implement ACO algorithms on a PC which has 4 GTX 480 GPUs. We implement two types of ACO models; the island model, and the other is the master/slave model. When we compare the island model and the master/slave model, the island model shows promising speedup values on class (iv) QAP instances. On the other hand, the master/slave model showed promising speedup values both on classes (i) and (iv) with large-size QAP instances.
June 21, 2013 by hgpu
