一种面向智能交通应用的海量空间数据存储及并行处理方法

摘要

文中以智能交通应用中海量空间数据存储为研究对象,使用基于HDSF的分布式存储金字塔模型对空间数据进行分块;引入基于空间数据分类和删除技术的副本存储策略,将同一类影像存储在同一节点或者相近节点,降低连续访问延时;引入Spark作为空间信息查询的并行处理工具;利用RDD的快速存储机制,减少查询过程对I/O的读写,提高查询效率.实验结果证明了文中方法的有效性.%Based on intelligent transportation applications in huge amounts of spatial data storage as the research object,this paper uses the pyramid model based on distributed storage of HDSF to block the spatial data, and introduces the technology based on spatial data classification and delete storage strategy, for which the same or similar type of images are stored in the same node to reduce the access latency in a row.By introducing a Spark as a parallel processing of the spatial information query tool,and using the RDD fast storage mechanism,can reduce the query process to read and write I/O,and improve the query efficiency.The experimental result shows the effectiveness of the proposed method.