• 中国计算机学会会刊
  • 中国科技核心期刊
  • 中文核心期刊

J4 ›› 2007, Vol. 29 ›› Issue (8): 61-63.

• 论文 • 上一篇    下一篇

一种基于sFlow的网络流量分析方法

张红林 王宏   

  • 出版日期:2007-08-01 发布日期:2010-06-02

  • Online:2007-08-01 Published:2010-06-02

摘要:

摘要:随着网络应用越来越广泛,关键的企业应用对网络的依赖性也越来越强,在网络使用上的一些很小的变化都有可能对网络的性能和可靠性造成很大的影响。因此,进行全方位、高速、实时的网络流量分析及网络性能分析成为一项有挑战性的、具有重要意义的研究课题。先前的流量监测技术(如RMON、NetFlow技术)都有许多缺点,而唯有s sFlow技术能够“看透”当今高速、复杂的网络。通过sFlow技术可以采集不同类别的大量的原始信息,但仅仅拥有这些原始信息还远远不够,还需要对这些原始信息进行分析析,从而得出对网络管理者有用的性能信息。本文提出了一种对利用sFlow技术采样的原始数据进行流量分析的方法,并给出了一个基于sF]ow的网络流量分析系统架构。

关键词: sFlow技术 系统架构 流量分析 性能分析

Abstract:

With the wider applications of the network, the key enterprise applications rely more and more on it. The small changes in the use of the network can  bring great impact on the performance and reliability of the network. So it is a challenge and important research topic to give a full-range, high-speed and real-time network traffic analysis and performance analysis. Previous traffic monitoring and measurement techniques (such as RMON, NetFlow technol logy) have many flaws, and only the sFlow technology can look through the modem high-speed and complex network. Using the sFlow technology can capture the various raw data,but it is not enough to get this raw information, and we have to analyze the raw data and obtain the performance information for t  he network manager from the raw data. This paper proposes a method to analyze the traffic using the sFlow sampled raw data,and gives a framework of traf fic analysis based on sFlow.

Key words: (sFlow technology, framework, traffic analysis, performance analysis)