Parallel Branch and Bound on a CPU-GPU System
CNRS; LAAS; 7 avenue du colonel Roche, F-31077 Toulouse, France
Euromicro International Conference on Parallel Distributed and Network-based Processing (PDP 2012), pp.392-398, 2012
@article{boukedjar2012parallel,
title={Parallel Branch and Bound on a CPU-GPU System},
author={Boukedjar, A. and Lalami, M.E. and El-Baz, D.},
year={2012}
}
Hybrid implementation via CUDA of a branch and bound method for knapsack problems is proposed. Branch and bound computations can be carried out either on the CPU or on the GPU according to the size of the branch and bound list, i.e. the number of nodes. Tests are carried out on a Tesla C2050 GPU. A first series of computational results showing a substantial speedup is displayed and analyzed.
March 13, 2012 by hgpu
