基于Spark的土地利用矢量数据空间叠加分析方法

Spatial overlay analysis of land use vector data based on Spark

针对海量土地利用矢量数据空间叠加分析的效率问题,提出了基于Spark的土地利用矢量数据空间叠加分析方法,通过弹性分布式数据集实现索引过滤与叠加计算,为解决空间叠加分析的瓶颈问题做了新的尝试.实验结果表明,相较基于Oracle数据管理的叠加分析方法,该方法显著提高了空间叠加分析效率,更适合面向海量土地利用矢量数据的叠加分析.

To increase the spatial overlay analysis efficiency of massive land use vector data,we proposes a method of land use vector data spatial overlay analysis based on Spark.The method realizes index filtering and overlay calculating of land use vector data by resilient distributed datasets（RDD）,which is a basic data structure of Spark.It makes a new attempt to overcome the bottleneck of land use vector data spatial overlay analysis.Comparing with the traditional overlay analysis based on Oracle data management,the approach increases the spatial overlay analysis efficiency significantly,which is more suitable for spatial overlay analysis with massive land use vector data,and can help us to manage and analyse the massive land use vector data.