A CUDA-enabled Parallel Implementation of Collaborative Filtering
School of Computer and Control, University of Chinese Academy of Sciences, Beijing 100190, China
Procedia Computer Science, Volume 30, Pages 66-74, 2014
@article{wang2014cuda,
title={A CUDA-enabled Parallel Implementation of Collaborative Filtering},
author={Wang, Zhongya and Liu, Ying and Ma, Pengshan},
journal={Procedia Computer Science},
volume={30},
pages={66–74},
year={2014},
publisher={Elsevier}
}
Collaborative filtering (CF) is one of the essential algorithms in recommendation system. Based on the performance analysis, two computational kernels are identified. In order to accelerate CF on large-scale data, a CUDA-enabled parallel CF approach is proposed where an efficient data partition scheme is proposed as well. Various optimization techniques are also applied to maximize the performance of the GPU. The experimental results demonstrate up to 48x speedup on a single Tesla C2070 graphics card.
June 1, 2014 by hgpu
