警务信息化系统中云存储平台的设计和实现

发布时间：2018-12-16 20:25

【摘要】：随着计算机的普及，网络技术的快速发展，警务系统开始利用计算机技术替代一部分的人工作业，以更好的发挥警务系统的作用。警务信息化系统为警务人员提供了从警务信息的登记录入到统计查询整个流程的功能。但是当前传统的警务系统在存储层面面临着一些局限：首先警务数据数量的快速增加，特别是在警务数据联网之后，大数据量增加了传统数据库的负担，其次警务系统的数据库直接建立在服务器之上，和其他程序共享资源，隔离性差，，再次警务系统的存储平台几乎没有扩展性，最后面对大数据，传统数据库难以进行存储和分析，特别是那些非结构化数据。因此本文尝试从存储的角度对警务信息化进行优化，主要包括以下三个部分： 1.将传统的数据库建立在新的存储平台之上。这个存储平台建立在数据库和服务器硬件中间，利用KVM虚拟化技术将底层的计算资源统一，然后利用CloudStack管理虚拟化之后的服务资源。通过创建虚拟机的方式为数据库提供平台。 2.利用Hadoop建立一个分布式文件系统，所有的大数据都可以存储在HDFS中，然后利用MapReduce进行并行化分析。利用Hadoop对大数据的支持，为警务系统提供包括警务数据、警务程序日志分析等功能。 3.利用Sqoop等工具，将关系型数据库中的数据导入HDFS，然后将HDFS中分析得到的结果数据再次导入关系型数据库中。以此来实现两个子系统之间的数据迁移功能。本文从存储的角度对警务信息化平台做了新的设计和实现，利用云计算的技术，为存储平台增加了可靠性、可扩展性和大数据能力。随着警务信息化的深入和发展，这种存储设计思路将体现出它的优势。
[Abstract]:With the popularization of computer and the rapid development of network technology, police service system began to use computer technology to replace part of the manual operation, in order to play a better role in the police service system. Police information system provides police officers with the function from the registration of police information to the whole process of statistical inquiry. However, the traditional police system is facing some limitations at the storage level: firstly, the number of police data increases rapidly, especially after the police data network, the large amount of data increases the burden of the traditional database. Secondly, the database of the police service system is built directly on the server, sharing resources with other programs, and the isolation is poor. Again, the storage platform of the police system has almost no expansibility. Finally, facing big data, Traditional databases are difficult to store and analyze, especially unstructured data. Therefore, this paper attempts to optimize the police informatization from the perspective of storage, mainly including the following three parts: 1. The traditional database is built on the new storage platform. The storage platform is based on database and server hardware, and unifies the underlying computing resources by using KVM virtualization technology, and then uses CloudStack to manage the service resources after virtualization. Provides a platform for the database by creating virtual machines. 2. Using Hadoop to build a distributed file system, all big data can be stored in HDFS, and then parallelized by MapReduce. With the support of Hadoop to big data, we can provide police service system with police data, police program log analysis and so on. 3. Using Sqoop and other tools, the data from relational database are imported into HDFS, and the result data from HDFS are imported into relational database again. The function of data migration between two subsystems is realized. In this paper, the police information platform is designed and implemented from the perspective of storage. The cloud computing technology is used to increase the reliability, expansibility and big data ability of the storage platform. With the deepening and development of police information, this storage design will reflect its advantages.
【学位授予单位】：电子科技大学
【学位级别】：硕士
【学位授予年份】：2013
【分类号】：TP333

【参考文献】