大规模GNSS网分布式存储与解算方法  

A Distributed Storage and Processing Method of Large-Scale GNSS Network

作　　者：李林阳[1] 崔阳[1] 陈正生 王宁[1] 吕志平[1] LI Linyang CUI Yang CHEN Zhengsheng WANG Ning LV Zhiping(Information Engineering University, Zhengzhou 450001, China Rocket Force University of Engineering, Xi＇ an 710025, China)

机构地区：[1]信息工程大学,河南郑州450001 [2]火箭军工程大学,陕西西安710025

出　　处：《测绘科学技术学报》2016年第5期464-469,共6页Journal of Geomatics Science and Technology

基　　金：国家自然科学基金项目(41274015);国家863计划项目(2013AA122501);信息工程大学地理空间信息学院硕士学位论文创新与创优基金项目(XS201501);地理信息工程国家重点实验室开放基金项目(SKLGIE2014-M-1-5)

摘　　要：针对GNSS大网在海量数据存储和解算方面面临的挑战,设计了从基础设施、数据管理、计算和服务到应用的GNSS分布式存储与解算体系结构;基于HDFS改进了GNSS数据存储目录结构;基于HBase设计了GNSS数据存储表结构;基于MapReduce提出了GNSS大网分区分布式解算策略。实验搭建了分布式平台,实现了GNSS数据分布式存储、并行检索和发布以及GNSS大网分区分布式解算,存储和解算性能均得到了较大提高。结果表明,所提出的方法可用于GNSS大网数据存储和解算。According to the challenge of storage and processing faced by large-scale GNSS network, the GNSS distributed storage and processing architecture from infrastructure, data management, computing and service to application is designed. Firstly, based on HDFS, the storage directory is improved. Secondly, based on HBase, the table architecture is designed. Thirdly, based on MapReduce, sub-network distributed processing is proposed. The distributed platform is built in the experiment, and the distributed storage, parallel retrial, publication and distributed processing are achieved. The performances of storage and computing are significantly improved. The result shows that the method proposed by this paper can be applied to the storage and computing of large network.

关 键 词：全球卫星导航系统 HADOOP平台 GNSS大网 分布式存储 分布式计算 

分 类 号：P228[天文地球—大地测量学与测量工程]