ACO on Multiple GPUs with CUDA for Faster Solution of QAPs
Hannan University, Matsubara Osaka 580-8502, Japan
Parallel Problem Solving from Nature (PPSN XII), 2012
@incollection{tsutsui2012aco,
year={2012},
isbn={978-3-642-32963-0},
booktitle={Parallel Problem Solving from Nature – PPSN XII},
volume={7492},
series={Lecture Notes in Computer Science},
editor={Coello, CarlosA.Coello and Cutello, Vincenzo and Deb, Kalyanmoy and Forrest, Stephanie and Nicosia, Giuseppe and Pavone, Mario},
doi={10.1007/978-3-642-32964-7_18},
title={ACO on Multiple GPUs with CUDA for Faster Solution of QAPs},
url={http://dx.doi.org/10.1007/978-3-642-32964-7_18},
publisher={Springer Berlin Heidelberg},
author={Tsutsui, Shigeyoshi},
pages={174-184}
}
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 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 on both classes (i) and (iv) with large-size QAP instances.
September 24, 2012 by hgpu
