采用回归方法优化网络流量管理模型处理性能

计算机工程与应用 ›› 2012, Vol. 48 ›› Issue (4): 104-106.

采用回归方法优化网络流量管理模型处理性能

邬平1，2，吴斌1，2

1.云南省科学技术情报研究院网络管理中心，昆明 650051
2.昆明理工大学计算机技术应用重点实验室，昆明 650051

收稿日期:1900-01-01 修回日期:1900-01-01 出版日期:2012-02-01 发布日期:2012-04-05

Optimizing process performance of network traffic management model using regression analysis

WU Ping1，2, WU Bin1，2

1.Network Administration Center, Yunnan Academy of Scientific & Technical Information, Kunming 650051, China
2.Computer Technology Application Key Laboratory, Kunming University of Science and Technology, Kunming 650051, China

Received:1900-01-01 Revised:1900-01-01 Online:2012-02-01 Published:2012-04-05

摘要/Abstract

摘要： 为探索优化网络流量管理性能的有效途径，在分析“深度报文检测和深度流行为检测”网络流量管理模型结构的基础上，确定了影响网络流量管理效率的性能指标及其计算方法，通过实际抽样流量数据计算性能指标值，建立性能指标关系散点图，发现性能指标间呈线性特征，采用多元回归方法建立性能指标估计函数，并利用标准化残差估计方法验证了函数的可用性和适应性，从而得到优化网络流量管理性能的定量计算方法。

关键词: 多元回归方法, 网络流量管理模型, 深度报文检测, 深度流行为检测, 线性回归函数

Abstract: For seeking the valuable method to optimize the performance of network traffic management, the performance indexes and calculation method are ascertained on the basis of analyzing the structure of the network traffic management model which composes of the deep packet inspection and the deep flow inspection. Through sampling traffic data to calculate performance indexes, the scatter plot of performance indexes is build up and the linear character is observed among indexes. Furthermore, the evaluated function is set up by using the multivariate regression method and validated by the standard residual function to prove the usability and adaptability. As a result, the quantum calculation method is acquired for optimizing the performance of network traffic management.

Key words: multivariate regression method, network traffic management model, deep packet inspection, deep flow inspection, linear regression function

邬平1，2，吴斌1，2. 采用回归方法优化网络流量管理模型处理性能[J]. 计算机工程与应用, 2012, 48(4): 104-106.

WU Ping1，2, WU Bin1，2. Optimizing process performance of network traffic management model using regression analysis[J]. Computer Engineering and Applications, 2012, 48(4): 104-106.