Dispatch strategy on massive 3D spatial data in geological space

Abstract

Three-dimensional visualization has become the development trend of geological information technology, which will inevitably lead to the sharp increase of geological spatial data. However, the inherent characteristics of underground geological data, such as invisible, non-parametric, unstructured and non-homogeneous, lead to great difference between underground geological data and ground surface data in the aspects of data representation, data storage and data application. Therefore the existing dispatch technology based on regular ground surface data structure, such as DOM or DEM, is not sufficient to ensure the efficient dispatch of underground geological spatial data. New underground geological data dispatch strategy is investigated in this paper, geological data can be loaded according to the following steps:(1) Spatial index is built and stored in the database after the geological data is inserted into spatial database, and trigger mechanism is introduced to maintenance the consistency of spatial index during the operation of data add, delete and update; (2) Connection pool technology is used to create several connections to spatial database when the client program is started, which makes preparation for multi-thread concurrent dispatch later, meanwhile spatial index is loaded into client memory; (3) In the process of data dispatch, client first retrieves spatial index according to user's query to get the index items of spatial objects, then client traverses the object cache to filter out objects which are not loaded into memory, finally multi-threading technology is started up for parallel dispatching spatial objects from spatial database;(4)Appropriate cache replacement algorithm is adopted to solve the lack of memory when the spatial objects are added into client memory; (5) In the interval of dispatch, such as the client is drawing or editing, pre-dispatch technology is used to assist data dispatch. At last, the dispatch strategy on massive 3D spatial data in geological space is created, which can provide 20-40% better performance than those methods existed already. This Strategy has been successfully applied in the project called “research on 3D spatial data management and analysis components” in national 863 program of China.