A Dynamic Pyramid Rtree Index for Massive GeoSpatial Files

Abstract

In the large-scale geography information system application, the distributed file system is the key approach for data storage and retrieving, but the traditional distributed spatial index's efficiency is low. We present a dynamic pyramid Rtree(DPR) index for the massive geospatial files management, the method is based on the pyramid spatial index and Rtree index, which can be adjusted with the data distributed dynamically, we described index building and data accessing, the experimentation result shows that our method's efficiency in distributed file system is higher than Hilbert tree index and Quad tree index in range query.