基于OpenMP的矢量空间数据并行拓扑算法设计与实现

Design and realization of parallel topology algorithm of vector spatial data based on OpenMP

高性能并行GIS逐渐成为GIS发展的新方向。矢量数据的复杂性使得一些并行GIS算法难以实现,从而无法满足并行GIS的发展要求。文中针对GIS算法中的拓扑算法,借助OpenMP编程模型,通过消除并行拓扑处理过程中的数据依赖,在单机多核的环境下设计并实现了矢量空间数据并行拓扑算法。通过实验对比串行拓扑算法和并行拓扑算法的处理时间和结果,验证了并行拓扑算法的正确性,同时证明并行拓扑算法能够在一定程度上提升拓扑处理的效率。

High performance parallel GIS has gradually become a new direction for the development of GIS . The complexity of vector data makes it difficult for a number of parallel GIS algorithm to be realized to satisfy the requirements of the development of parallel GIS .T his paper focuses on one of GIS algorithms—the topological algorithm .Relying on OpenM P ,it can achieve that parallel topology algorithm for spatial vector data is designed and implemented under single‐machine and multi‐core environment ,by eliminating data dependence of the parallel topological processing .Via experimental comparison of the result and the processing time between serial topology algorithm and parallel topology algorithm processing ,the paper also verifies the correctness of the parallel topological algorithm and proves that the parallel algorithm can promote the efficiency of topology processing to a certain extent .