基于非关系型数据库集群的车辆监控系统设计与实现

发布时间：2018-06-04 06:01

本文选题：非关系型数据库 + MongoDB　；参考：《北京邮电大学》2016年硕士论文

【摘要】：随着互联网的飞速发展与普及,越来越多基于位置的服务(Location Based Service,LBS)的应用出现在我们的生活中。车辆监控系统就是一种典型的针对特定职能部门的LBS应用。随着系统中数据量的飞速增长,如何高效地存储、处理海量的数据,并保持系统在高并发负载时的性能已成为一个亟待解决的问题。在传统的关系型数据库面对“大数据”、“高可用性”、“高扩展性”等问题捉襟见肘时,非关系型数据库(NotOnly SQL,NoSQL)异军突起,成为了一个热门的研究领域。本文结合车辆监控系统的需求和特点,采用NoSQL数据库集群作为数据存储,设计并实现了一个能够高效存储海量数据且具有高扩展性的车辆监控系统,具有很强的实践意义。本文主要工作有:(1)介绍了 NoSQL数据库的概念、发展研究现状和理论基础,阐述了 NoSQL数据库在高并发负载、海量数据存储、高可用性和扩展性方面对传统的关系型数据库的优势,分析了为什么选择MongoDB作为系统的数据存储。(2)深入研究了 MongoDB的分片技术和负载均衡原理,针对车辆监控系统中的数据特点,设计了相应数据集合的分片策略。(3)部署搭建了 MongoDB数据库集群,并以此为基础,设计了数据库存储模型,结合Node.js、WebGIS等相关技术设计并实现了一个能够高效存储海量数据并具有高扩展性的车辆监控系统。(4)模拟真实环境,以单机MongoDB数据库为基准,对MongoDB集群进行了性能测试,验证了基于MongoDB集群的车辆监控系统具有更高的吞吐量和满足高并发负载的性能。
[Abstract]:With the rapid development and popularization of the Internet, more and more location-based Based Based (LBSs) applications appear in our lives. Vehicle monitoring system is a typical LBS application for specific functional departments. With the rapid growth of the data in the system, how to store and process the massive data efficiently and maintain the performance of the system under high concurrent load has become an urgent problem to be solved. When the traditional relational database is facing the problems of "big data", "high availability" and "high scalability", the non-relational database has become a hot research field. According to the requirements and characteristics of vehicle monitoring system, this paper uses NoSQL database cluster as data storage, and designs and implements a vehicle monitoring system which can store large amounts of data efficiently and has high expansibility, which has strong practical significance. The main work of this paper is to introduce the concept of NoSQL database, research status and theoretical basis, and explain the high concurrent load and mass data storage of NoSQL database. The advantages of high availability and extensibility to traditional relational databases are analyzed. Why MongoDB is chosen as the data storage of the system. (2) the slicing technology and load balancing principle of MongoDB are deeply studied. According to the data characteristics of the vehicle monitoring system, the partition strategy of the corresponding data set is designed and the MongoDB database cluster is built. Based on this, the database storage model is designed. This paper designs and implements a vehicle monitor and control system, which can store massive data efficiently and has high expansibility, and simulates the real environment with Node.jsGIS-WebGIS. The performance of MongoDB cluster is tested based on single MongoDB database. It is verified that the vehicle monitoring system based on MongoDB cluster has higher throughput and higher concurrent load performance.
【学位授予单位】：北京邮电大学
【学位级别】：硕士
【学位授予年份】：2016
【分类号】：TP311.13;TP277

【参考文献】