Network abnormal behavior in the cloud data center not only causes serious traffic load on the network equipment

but also greatly reduces the degree of user experience.The shared resource mode in the cloud computing environment and the very different business forms of users make that network analysis and abnormal behavior orientation are extremely difficult.In this paper

we extract and analyze the characteristics of network anomaly behavior in cloud data center

and deeply analyze the network architecture and principle of SDN cloud data center.We summarize the method of network anomaly behavior determination based on OpenFlow flow-table.Moreover

this paper puts forward an intelligent system for automated detection and blocking of network abnormal behavior to realize fast processing of network anomaly behavior in the help of automated operation and maintenance.