基于Apache Storm的大规模网络流量实时监控系统研究  

Research of Apache Storm based real-time traffic monitoring system for large-scale network

作　　者：姚欣[1] 王劲松[1] YAO Xin WANG Jin-song(School of Computer and Communication Engineering, Key Laboratory of Intelligent Computing and Novel Software Technology, Tianjin University of Technology, Tianjin 300384, Chin)

机构地区：[1]天津理工大学计算机与通信工程学院天津市智能计算与软件新技术重点实验室,天津300384

出　　处：《天津理工大学学报》2016年第6期25-29,47,共6页Journal of Tianjin University of Technology

基　　金：国家自然科学基金(61272450)

摘　　要：近些年来,接入设备的激增、网络应用及协议的不断涌现和互联网流量的爆发式增长为数据分析与处理带来极大挑战,计算速度以及横向扩展能力已成为制约传统架构模式的网络流量监控系统向高并发网络环境迁移的主要瓶颈.本文面向大规模网络,提出并实现了一种以Storm Topology计算模型为基础的网络流量实时监控系统设计方案.通过对天津教育城域网实时Netflow数据的采集与处理,本文对该设计方案的计算性能进行了验证,此外,本文还从系统吞吐量和并发节点数对计算速度的影响这两方面对Storm集群进行了分析与探究.It has been a great challenge to processing data in recent years due to the significant increasing of terminal devices, web applications, protocols and network traffic. Processing latency and extensibility are the main restrictions of the network traffic monitoring systems that follows traditional system architecture. In this article we propose a Storm Topology computation model based traffic monitoring system for large-scale network. We evaluate our system by the real-time Netflow data of Tianjin Education and Research Network （TERNET）.

关 键 词：大规模网络 实时监控 STORM NETFLOW 

分 类 号：TP311.563[自动化与计算机技术—计算机软件与理论]