移动互联网流量数据的高效处理与性能优化
发布时间：2019-01-24 08:17
【摘要】：随着科技的快速发展,网络已经成为了人们不可或缺的工具,网络的应用种类也越来越丰富。同时,移动通信技术的飞速发展,导致全球的移动通信技术开始从2G向3G演进。手机网民数量已超过传统PC端的网民数量,而且以更快的速度不断地增长。 本文首先介绍网络流量监控技术和移动互联网络流量数据特征,接下来介绍了Google核心云计算技术以及该技术的一个完整的实现框架——Hadoop平台。 为了更加高效地处理流量数据,本文设计实现了基于Hadoop的分布式数据处理框架(Hadoop-Based Distributed Data Process System, HDPS),着重分析了HDPS数据分析模块的架构设计与实现方法。同时,本文介绍了基于HDPS研发的几个应用系统——省网监测系统和高性能分布式移动网络流量数据存储查询系统,并着重探讨高性能分布式移动网络流量数据存储查询系统的设计和实现。 最后,本文总结了分布式平台性能优化的经验和方法,并介绍了这些性能优化方式的应用场景。
[Abstract]:With the rapid development of science and technology, the network has become an indispensable tool. At the same time, with the rapid development of mobile communication technology, the global mobile communication technology began to evolve from 2G to 3G. The number of mobile phone users has exceeded the number of traditional PC users, and growing at a faster pace. This paper first introduces the network traffic monitoring technology and the characteristics of mobile Internet traffic data, then introduces the core cloud computing technology of Google and a complete implementation framework of this technology, Hadoop platform. In order to deal with traffic data more efficiently, this paper designs and implements a distributed data processing framework based on Hadoop (Hadoop-Based Distributed Data Process System, HDPS), focuses on the analysis of HDPS data analysis module architecture design and implementation method). At the same time, this paper introduces several application systems based on HDPS, which are the provincial network monitoring system and the high performance distributed mobile network traffic data storage and query system. The design and implementation of high performance distributed mobile network traffic data storage and query system are discussed. Finally, this paper summarizes the experience and methods of performance optimization of distributed platform, and introduces the application scenarios of these performance optimization methods.
【学位授予单位】：北京邮电大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：TP393.06;TN929.5
本文编号：2414289
[Abstract]:With the rapid development of science and technology, the network has become an indispensable tool. At the same time, with the rapid development of mobile communication technology, the global mobile communication technology began to evolve from 2G to 3G. The number of mobile phone users has exceeded the number of traditional PC users, and growing at a faster pace. This paper first introduces the network traffic monitoring technology and the characteristics of mobile Internet traffic data, then introduces the core cloud computing technology of Google and a complete implementation framework of this technology, Hadoop platform. In order to deal with traffic data more efficiently, this paper designs and implements a distributed data processing framework based on Hadoop (Hadoop-Based Distributed Data Process System, HDPS), focuses on the analysis of HDPS data analysis module architecture design and implementation method). At the same time, this paper introduces several application systems based on HDPS, which are the provincial network monitoring system and the high performance distributed mobile network traffic data storage and query system. The design and implementation of high performance distributed mobile network traffic data storage and query system are discussed. Finally, this paper summarizes the experience and methods of performance optimization of distributed platform, and introduces the application scenarios of these performance optimization methods.
【学位授予单位】：北京邮电大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：TP393.06;TN929.5
【参考文献】
相关期刊论文 前2条
1 侯捷;;Java反射机制[J];程序员;2004年10期
2 唐兴;;移动通信技术的历史及发展趋势[J];江西通信科技;2008年02期
，本文编号：2414289
本文链接：https://www.wllwen.com/guanlilunwen/ydhl/2414289.html
