GPU Algorithms for the Estimation of Environmental Models Based on Large Datasets
Department of Information Technology and Mathematical Methods, University of Bergamo, viale Marconi, 5 – 24044 – Dalmine, Italy
University of Bergamo, 2011
@article{finazzi2011gpu,
title={GPU ALGORITHMS FOR THE ESTIMATION OF ENVIRONMENTAL MODELS BASED ON LARGE DATASETS},
author={Finazzi, F. and Cameletti, M.},
year={2011}
}
Statistical environmental models are computationally intensive due to the high dimension of the data, both in space and time, and due to the inferential techniques required for parameter estimation and spatial prediction. In particular, the complexity of these procedures is related to matrix operations (inversion, solution of linear systems, factorization) involving large matrices. Recently, much attention has been paid around the possibility of taking advantage of graphics processing units (GPUs) for mathematical computation. The GPUs provide a high degree of parallelism at a reasonable cost and may represent a viable alternative compared to the classical cluster con?gurations. In this work, we develop the shared library GPU4GEO implementing ad-hoc linear-algebra functions running on GPUs and compare them with the standard algorithms for CPU. As an example, we apply the GPU algorithms to estimate the parameters of a non-separable space-time model for air quality data.
December 15, 2011 by hgpu
