Parallel GPU Implementation of Iterative PCA Algorithms
Institute for Biocomplexity and Informatics, University of Calgary, 2500 University Drive NW, Calgary, Alberta, T2N 1N4, Canada
Journal of Computational Biology. November 2009, 16(11): 1593-1599, arXiv:0811.1081 [q-bio.QM] (7 Nov 2008)
@article{andrecut2009parallel,
title={Parallel gpu implementation of iterative pca algorithms},
author={Andrecut, M.},
journal={Journal of Computational Biology},
volume={16},
number={11},
pages={1593–1599},
issn={1066-5277},
year={2009},
publisher={Mary Ann Liebert, Inc. 140 Huguenot Street, 3rd Floor New Rochelle, NY 10801 USA}
}
Principal component analysis (PCA) is a key statistical technique for multivariate data analysis. For large data sets the common approach to PCA computation is based on the standard NIPALS-PCA algorithm, which unfortunately suffers from loss of orthogonality, and therefore its applicability is usually limited to the estimation of the first few components. Here we present an algorithm based on Gram-Schmidt orthogonalization (called GS-PCA), which eliminates this shortcoming of NIPALS-PCA. Also, we discuss the GPU (Graphics Processing Unit) parallel implementation of both NIPALS-PCA and GS-PCA algorithms. The numerical results show that the GPU parallel optimized versions, based on CUBLAS (NVIDIA) are substantially faster (up to 12 times) than the CPU optimized versions based on CBLAS (GNU Scientific Library).
November 13, 2010 by hgpu
