Implementation of the r.cuda.los module in the open source GRASS GIS by using parallel computation on the NVIDIA CUDA graphic cards
Telekom Slovenije d.d., Cigaletova 15, 1000 Ljubljana, Slovenija
Elektrotehniski Vestnik, 79(1-2), 19-24, 2012
@article{osterman2012implementation,
title={Implementation of the r.cuda.los module in the open source GRASS GIS by using parallel computation on the NVIDIA CUDA graphic cards},
author={Osterman, A.},
journal={ELEKTROTEHNISKI VESTNIK},
volume={79},
number={1-2},
pages={19–24},
year={2012}
}
Parallel computing is in expanding phase in GIS applications. A very attractive solution for parallel computing are the NVIDIA graphic cards, with a parallel computing platform and the CUDA (Compute Unified Device Architecture) programming model. The basis for this paper is the r.los module used to calculate optical visibility (LOS – Line of Sight), which is already implemented in the GRASS GIS environment. A completely new r.cuda.los module with the same functionality as the r:los module is presented. By using the r.cuda.los module for radio planning purposes of limiting the computation along the vertical and horizontal angle is also make possible. Visibility is calculated for each slice. The responsibility for the calculation of each slice is with its own thread from the parallel processor. At the size of the map of 28161 x 17921 points with the resolution 12.5m x 12.5m, the computation time is 18 s. In parallel computing the GIS data, the performance can be one, two or even three size classes faster than in the sequential computing.
July 20, 2012 by hgpu
