【摘要】 以实现数据中心端口故障检测为目的，设计基于SDN的数据中心端口故障检测方法。该方法利用WIN32平台的网络分析和数据包捕获链接库WinCap捕获基于SDN网络体系的数据中心组网端口通信协议数据包，并通过t-SEN算法对其展开降维处理，再使用多尺度排列熵算法提取降维后的通信协议数据包特征；利用支持向量机算法设计数据中心端口故障检测引擎，并使用梯度算法对支持向量机参数实施优化后，将通信协议数据包特征输入到数据中心端口故障检测引擎内，启动该引擎输出数据中心端口故障检测结果。实验结果表明：该方法可有效对通信协议数据包进行降维处理并保留其自身特征，同时在不同数据包大小和相似度情况下，其提取数据包特征的支持度均高于90%;并有效检测不同类型的数据中心端口故障。更多还原

【Abstract】 In order to realize data center port fault detection, a data center port fault detection method based on SDN is designed. In this method, the network analysis and packet capture link library WINCAP of Win32 platform is used to capture the communication protocol packet of the networking port of the data center based on the SDN network system, and the dimensionality is reduced by t-sen algorithm, and then the dimensionality reduced communication protocol packet characteristics are extracted by multi-scale permutation entropy algorithm; The support vector machine algorithm is used to design the data center port fault detection engine, and the gradient algorithm is used to optimize the parameters of the support vector machine. After inputting the characteristics of the communication protocol packet into the data center port fault detection engine, the engine is started to output the data center port fault detection results. The experimental results show that this method can effectively reduce the dimension of communication protocol packets and retain their own characteristics. At the same time, the support of extracting packet features is higher than 90% under different packet sizes and similarities; And effectively detect different types of data center port failures.更多还原