非关系型与关系型空间数据库对比分析与协同应用研究

发布时间：2018-05-07 09:54

本文选题：空间数据库 + 空间数据管理　；参考：《首都师范大学》2013年硕士论文

【摘要】：近年来随着遥感技术、地理信息系统技术的不断发展,GIS的应用领域越发广泛,GIS中空间数据量呈指数级别增长,尤其是大数据和非结构化数据的大量出现,对常用的关系型或对象关系型数据库的空间数据管理能力提出了巨大挑战。随着Web2.0时代的到来,为满足爆炸式增长的互联网的应用需求,非关系型数据库这一全新的数据管理技术应运而生。非关系型数据库在存储、查询和管理大数据和非结构化数据方面具有优势。本文对关系型和非关系型空间数据库进行对比研究,从中选出MongoDB、ArcSDE(Oracle)和PostGIS作为代表,从体系架构、存储机制和操作方式等方面进行了分析,通过选定定性和定量评价因子建立评价体系,并开展相关实验,指出非关系和关系型数据空间数据库空间数据管理的特点；在此基础上,设计了非关系型和关系型数据库协同应用方案,并基于OGC的简单要素模型和面向对象和组件技术,对其中的关键技术进行实现；将协同应用方案在地震救援数据管理系统中进行了应用并对结果进行了评价分析。本文主要的研究结论和成果如下： 1.通过建立的定量和定性的评价指标体系,对MongoDB、ArcSDE和PostGIS进行对比研究,得出在相同数据大小下MongoDB存储空间矢量数据较ArcSDE和PostGIS消耗时间少,而存储相同大小栅格数据ArcSDE的速率最快,且磁盘消耗少：在空间查询时,相同数据范围MongoDB的查询效率高于其他两者,范围越大,优势越明显。 2.根据对比分析的结果,提出MongoDB存储管理矢量空间数据和空间元数据,ArcSDE存储管理栅格数据的协同应用方案；并对协同中的非关系型数据库空间数据存储,关系型和非关系型空间数据库存储数据的迁移和采用统一管理方式对两者进行管理的关键技术进行了实现。 3.根据地震救援数据管理的需求,将非关系型数据库和关系型数据库协同方案应用于地震救援数据管理系统中。采用MongoDB和ArcSDE协同应用对系统进行了设计和实现,实现结果表明非关系型和关系型数据库协同提高了地震救援数据的存储和查询效率,降低了数据存储的磁盘消耗,且使得数据库的更新和扩展变的更加方便。能更好地满足地震救援数据管理的需求。
[Abstract]:In recent years, with the development of remote sensing technology and the continuous development of GIS technology, the spatial data volume in GIS is increasing exponentially, especially the emergence of big data and unstructured data. This paper presents a great challenge to the spatial data management ability of relational or object-relational databases. With the arrival of Web2.0 era, in order to meet the demand of explosive growth of Internet application, non-relational database, a new data management technology, came into being. Non-relational databases have advantages in storing, querying and managing big data and unstructured data. This paper makes a comparative study of relational and non-relational spatial databases, and selects MongoDB ArcSDE Oracle and PostGIS as representatives to analyze the architecture, storage mechanism and operation mode of the database. By selecting qualitative and quantitative evaluation factors, the evaluation system is established, and relevant experiments are carried out. The characteristics of spatial data management in non-relational and relational data spatial databases are pointed out. The cooperative application scheme of non-relational and relational database is designed, and the key technologies are implemented based on the simple element model of OGC and object-oriented and component technology. The cooperative application scheme is applied to the seismic rescue data management system and the results are evaluated and analyzed. The main conclusions and results of this paper are as follows: 1. The quantitative and qualitative evaluation index system was established to compare MongoDBE ArcSDE and PostGIS. The results show that the storage time of MongoDB space vector data is less than that of ArcSDE and PostGIS under the same data size, but the rate of storing the same size raster data ArcSDE is the fastest. And the disk consumption is less: in spatial query, the query efficiency of MongoDB in the same data range is higher than that of the other two, the larger the scope, the more obvious the advantage. 2. According to the results of comparative analysis, this paper proposes a collaborative application scheme of MongoDB storage and management vector spatial data and spatial metadata storage and management raster data, and stores spatial data of non-relational database in collaboration. The key technologies of storing data in relational and non-relational spatial databases and using unified management are implemented. 3. According to the requirement of seismic rescue data management, the cooperative scheme of non-relational database and relational database is applied to the seismic rescue data management system. The system is designed and implemented by the cooperative application of MongoDB and ArcSDE. The results show that non-relational and relational databases can improve the storage and query efficiency of seismic rescue data and reduce the disk consumption of data storage. And make the update and expansion of the database more convenient. It can better meet the needs of earthquake rescue data management.
【学位授予单位】：首都师范大学
【学位级别】：硕士
【学位授予年份】：2013
【分类号】：P208

【参考文献】