On Massive Spatial Data Retrieval Based on Spark

In order to search more efficiently for rapidly growing spatial data, cloud quad tree and R-tree is adopted in spatial index for the non-relational databases of cloud HBase, by which data can be retrieved successfully. By comparison of retrieval efficiency between cloud quad tree and R-tree, that how different parameters acted is tested on data index and retrieval efficiency and put forward an relatively more reasonable solution. Subsequently, we verify the validity of index calculation when there is a giant one.