CUDA-Accelerated HD-ODETLAP: Lossy High Dimensional Gridded Data Compression
Rensselaer Polytechnic Institute, 110 8th Street, Troy, NY, USA 12180
2012 International Workshop on Modern Accelerator Technologies for GIScience (MAT4GIScience 2012), 2012
We present High-dimensional Overdetermined Laplacian Partial Differential Equations (HD-ODETLAP), a high dimensional lossy compression algorithm and CUDA implementation that exploits data correlations across multiple dimensions of gridded GIS data. Exploiting the GPU gives a considerable speedup. In addition, HD-ODETLAP compresses much better than JPEG2000 and 3D-SPIHT, when fixing either the average or the maximum error.
August 8, 2012 by hgpu
