作者投稿和查稿 主编审稿 专家审稿 编委审稿 远程编辑

计算机工程 ›› 2008, Vol. 34 ›› Issue (6): 112-114. doi: 10.3969/j.issn.1000-3428.2008.06.041

• 网络与通信 • 上一篇    下一篇

大规模网络中IP流长分布统计模型

吴 桦,周明中,龚 俭   

  1. (东南大学计算机科学与工程学院江苏省计算机网络技术重点实验室,南京 210096)
  • 收稿日期:1900-01-01 修回日期:1900-01-01 出版日期:2008-03-20 发布日期:2008-03-20

IP Flows Length Statistical Distribution Model in Large-scale Networks

WU Hua, ZHOU Ming-zhong, GONG Jian   

  1. (School of Computer Science and Engineering, Jiangsu Province Key Lab of Computer Networking Technology, Southeast Univ., Nanjing 210096)

  • Received:1900-01-01 Revised:1900-01-01 Online:2008-03-20 Published:2008-03-20

摘要: 对具有不同特征的大规模高速网络的TRACE进行分析,发现不同IP流的流长分布特征。在此基础上,提出大规模网络状况下IP流长分布经验模型,该模型在表达大规模网络IP流长分布上,其精度高于原有Pareto模型,复杂度低于原有双Pareto模型。对相关模型与实际TRACE流长分布的拟合程度进行了检验,并对模型的相关参数取值范围及该经验模型与现有模型存在的异同做了讨论,进而分析导致异同的原因,并指出IP流长分布的发展趋势。

关键词: 网络被动测量, 网络行为, 大规模网络, IP流长, 分布统计模型

Abstract: This paper studies TRACEs with different characteristics. The flows length distribution characteristics are discovered by analyzing those TRACEs. Based on the discussions, the empirical model of IP flows distribution for large-scale networks is proposed based on the characteristics analysis, whose precision is better than Pareto model, and the complexity is less than double Pareto model. Goodness-of-fit test is employed to inspect the effect of this model and its parameters. And then, this empirical model is contrasted with other distribution models presented by former researchers, and the same and different characteristics among all of these models are discussed, and so do their causes. The possible tendency of IP flow distribution is forecasted based on those discussions.

Key words: network passive measurement, network behavior, large-scale networks, IP flows length, statistical distribution model

中图分类号: