基于异构数据库的空间天文卫星数据组织方法  

Data Organization Method of Space Astronomical Satellite Based on Heterogeneous Database

作　　者：杨晓艳 刘志奇[1] 孙小涓 石涛 纪珍 佟继周[4] Yang Xiaoyan;Liu Zhiqi;Sun Xiaojuan;Shi Tao;Ji Zhen;Tong Jizhou(Aerospace Information Research Institute,Chinese Academy of Sciences,Beijing 100190,China;Key Laboratory of Technology in Geospatial Information Processing and Application System,Chinese Academy of Sciences,Beijing 100190,China;University of Chinese Academy of Sciences,Beijing 100049,China;National Space Science Center,Chinese Academy of Sciences,Beijing 100190,China)

机构地区：[1]中国科学院空天信息创新研究院,北京100190 [2]中国科学院空间信息与应用系统重点实验室,北京100190 [3]中国科学院大学,北京100049 [4]中国科学院国家空间科学中心,北京100190

出　　处：《天文研究与技术》2022年第5期500-508,共9页Astronomical Research & Technology

基　　金：中国科学院战略性先导科技专项空间科学(二期)科学卫星地面支撑项目;空间科学卫星大数据应用承载平台项目。

摘　　要：随着空间天文卫星获取的数据量越来越多,数据应用逐渐发挥了重要作用。现有的天文卫星地面系统中,数据存储方式和组织方法各异,数据量达PB级,并且数据量持续增长,无法快速查找并提取特征参数,难以满足数据应用对查询的时效性要求。提出了一种新的空间天文卫星数据组织方法,通过解析抽取数据中的海量特征参数,建立观测时间、空间位置与特征参数的关联,实现在统一时空下的多源数据组织;同时采用关系型数据库与非关系型数据库结合的异构存储方式,设计了海量特征参数存储管理系统。将本文方法应用于空间科学卫星大数据应用平台系统,使用硬X射线调制望远镜卫星数据的实验结果表明,系统能够较好地满足针对按照时间、空间条件获取数据的要求,相比关系型数据库的数据组织方式,相同查询模式下数据检索效率明显提高,并且随着数据存储量的增加,系统具有稳定的扩展能力。With the increase of the space astronomical satellite data volume, data applications have gradually played an important role. Data applications such as research on space astronomical targets, payload status monitoring, spatial target analysis, detection plan assistance formulation, detection process visualization, etc., all require the processing and analysis of multi-source massive data such as detection target information, satellite platform parameters, and payload parameters. In order to realize the rapid extraction of characteristic data, this paper proposes a new method of data organization for space detection satellites. The method extractes massive characteristic parameters from data files, and establishes the association of observation time, spatial location and characteristic parameters, so as to realize multi-source data organization under a unified time and space frame. Then, using the heterogeneous storage method of the combination of SQL database and No-SQL database, a characteristic parameter storage management system of 10 billion or even 100 billion is designed. The method in this paper is applied to the space science satellite big data application platform system. The experimental results of using HXMT(hard X-ray modulation telescope) satellite data show that the system can nicely meet the requirements of obtaining data according to time and space conditions;compared with relational database data organization method, the data retrieval efficiency under the same query mode is significantly improved;and with the increase of data storage capacity, the system has stable expansion capabilities.

关 键 词：空间天文数据 时空索引 数据组织 非关系型数据库 

分 类 号：P171.3[天文地球—天文学]