Parallelization of Regional Operation Algorithm Using Parallel Raster-based Geocomputation Operators
AI Beibe · Geo-information Science · 2015
Parallel raster-based programming libraries have been proposed to make the details of parallel programming and the parallel hardware architecture to be transparent to users in some degrees. Thus these libraries can facilitate the development of parallel programs of raster-based geocomputation. Among the existing parallel programming libraries, parallel raster-based geocomputation operators(Pa RGO), which is recently proposed by Qin et al, shows great advantages. This is not only because Pa RGO encapsulates the general steps in parallel raster-based geocomputation, but also because Pa RGO is compatible with multiple commonly used parallel computing platforms. Currently, Pa RGO is designed for supporting local operation, focal operation and global operation directly. However, the availability of Pa RGO for supporting regional operation in raster-based geocomputation has not been evaluated. In this paper, we evaluate Pa RGO to testify its performance in this circumstance by using a multiple-flow-direction algorithm as a representation of the regional operation. Different versions of Pa RGObased parallel programs for this algorithm are tested on a symmetrical multiprocessing(SMP) cluster and evaluated from two aspects: the performability and the parallel efficiency. The experimental results show that the current Pa RGO cannot directly support the parallelization of regional operations. But it can be supportive when the regional operation is transformed into an iteration process of focal operation. On a SMP cluster, MPI-version parallel program performs better than MPI/Open MP-version parallel program.