基于Hypertable空间数据库设计及应用
发布时间：2018-09-06 08:58
【摘要】：随着地理科学和计算机科学的不断发展,地理信息系统(GIS)的应用领域也在不断扩展。空间数据的存储是GIS系统的组织基础,空间数据的存取效率在很大程度上影响着GIS系统的性能。目前GIS系统对空间数据的组织大多基于两种形式：一种是借助于空间数据库引擎将空间数据存入关系型数据库中,一种是将空间数据存入特定格式的文件中,利用文件系统对空间数据进行管理。但这两种存储方案都存在一定的不足：空间数据库引擎将数据存储在关系型数据库中,关系型数据库在应对海量空间数据时存取性能不足,而且存储结构不够灵活；文件格式的空间数据操作较繁琐,空间数据存储在文件中也导致无法对这些数据进行统一的管理。 针对上述两种存储方案的不足,本文引入了Hypertable数据库来存储空间数据。Hypertable是基于Bigtable设计,以C++语言实现的非关系型分布式数据库,支持key-value数据模型。本文对基于Hypertable的空间数据库进行了详细的设计,将其应用到实际项目中,基于实验结果对其性能进行了评估,并与前述两种存储方案的性能进行了比较。本文的主要工作如下： 1)常见空间数据存储方案的研究及性能测试。研究了Oracle Spatial空间数据库引擎和NetCDF文件存储格式,并在项目实现时对两种存储方案进行了性能测试。 2)基于Hypertable空间数据库的设计。深入研究Hypertable数据库适合存储空间数据的特点以及在Hypertable上建立空间索引的方法。 3)设计并实现海浪气象信息动态展示系统。对海浪动态展示系统进行详细的设计与实现,并且在该系统中加入了基于浪高级别的航线安全评估模块。
[Abstract]:With the development of geography science and computer science, the application field of (GIS) is also expanding. Spatial data storage is the organizational basis of GIS system. The efficiency of spatial data access greatly affects the performance of GIS system. At present, the organization of spatial data in GIS system is mostly based on two forms: one is to store spatial data in relational database by means of spatial database engine, the other is to store spatial data in files in a specific format. The file system is used to manage the spatial data. But these two storage schemes have some shortcomings: the spatial database engine stores the data in the relational database, the relational database has insufficient access performance when dealing with the massive spatial data, and the storage structure is not flexible enough; The operation of spatial data in file format is more complicated, and the storage of spatial data in the file also leads to the difficulty of unified management of these data. In view of the shortcomings of the above two storage schemes, this paper introduces Hypertable database to store spatial data. Hypertable is a non-relational distributed database based on Bigtable, which is implemented in C language and supports the key-value data model. In this paper, the spatial database based on Hypertable is designed in detail, and applied to the actual project. The performance of the spatial database is evaluated based on the experimental results, and compared with the performance of the two storage schemes mentioned above. The main work of this paper is as follows: 1) Research and performance test of common spatial data storage scheme. This paper studies the Oracle Spatial spatial database engine and NetCDF file storage format, and tests the performance of the two storage schemes when the project is implemented. 2) Design based on Hypertable spatial database. The characteristics of Hypertable database suitable for storing spatial data and the method of establishing spatial index on Hypertable are studied. 3) the dynamic display system of wave weather information is designed and implemented. The dynamic wave display system is designed and implemented in detail, and the route safety evaluation module based on wave level is added to the system.
【学位授予单位】：北京邮电大学
【学位级别】：硕士
【学位授予年份】：2013
【分类号】：P208
本文编号：2225848
[Abstract]:With the development of geography science and computer science, the application field of (GIS) is also expanding. Spatial data storage is the organizational basis of GIS system. The efficiency of spatial data access greatly affects the performance of GIS system. At present, the organization of spatial data in GIS system is mostly based on two forms: one is to store spatial data in relational database by means of spatial database engine, the other is to store spatial data in files in a specific format. The file system is used to manage the spatial data. But these two storage schemes have some shortcomings: the spatial database engine stores the data in the relational database, the relational database has insufficient access performance when dealing with the massive spatial data, and the storage structure is not flexible enough; The operation of spatial data in file format is more complicated, and the storage of spatial data in the file also leads to the difficulty of unified management of these data. In view of the shortcomings of the above two storage schemes, this paper introduces Hypertable database to store spatial data. Hypertable is a non-relational distributed database based on Bigtable, which is implemented in C language and supports the key-value data model. In this paper, the spatial database based on Hypertable is designed in detail, and applied to the actual project. The performance of the spatial database is evaluated based on the experimental results, and compared with the performance of the two storage schemes mentioned above. The main work of this paper is as follows: 1) Research and performance test of common spatial data storage scheme. This paper studies the Oracle Spatial spatial database engine and NetCDF file storage format, and tests the performance of the two storage schemes when the project is implemented. 2) Design based on Hypertable spatial database. The characteristics of Hypertable database suitable for storing spatial data and the method of establishing spatial index on Hypertable are studied. 3) the dynamic display system of wave weather information is designed and implemented. The dynamic wave display system is designed and implemented in detail, and the route safety evaluation module based on wave level is added to the system.
【学位授予单位】：北京邮电大学
【学位级别】：硕士
【学位授予年份】：2013
【分类号】：P208
【参考文献】
相关期刊论文 前10条
1 刘文军;李靖;袁昌洪;刘方;解令运;;基于NetCDF数据模型的气象资料存储设计[J];安徽农业科学;2009年03期
2 马林波;张洪文;朱李忠;金丽华;宿裕;;基于Geodatabase极地NetCDF验潮数据的存储与分析[J];测绘工程;2012年02期
3 李莉莎;;关于NOSQL的思考[J];中国传媒科技;2010年04期
4 李振华;ORACLE空间数据库的对象-关系模式初探——兼议关系数据库和面向对象数据库在GIS中的应用[J];地球科学;2002年03期
5 陆丽珍;朱光良;刘仁义;;分布式栅格空间数据库引擎设计与实现[J];地球信息科学;2007年04期
6 黄贤立;;NoSQL非关系型数据库的发展及应用初探[J];福建电脑;2010年07期
7 程俊;杨卓宁;费江涛;;分布式海量数据管理系统Hypertable底层存储结构分析[J];兵工自动化;2010年07期
8 李滨,王青山,冯猛;空间数据库引擎关键技术剖析[J];测绘学院学报;2003年01期
9 张成才,孙喜梅,黄慧;SDE的实体-关系模型空间数据管理方式研究[J];计算机工程与应用;2003年02期
10 黄嵘;同淑荣;;使用Oracle Spatial进行空间数据建模研究[J];计算机工程与应用;2010年07期
相关硕士学位论文 前1条
1 李杰;海洋数据共享平台关键技术研究与开发[D];天津大学;2008年
，本文编号：2225848
本文链接：https://www.wllwen.com/kejilunwen/dizhicehuilunwen/2225848.html
