Design And Implementation Of Traffic Monitoring In Cloud Environment Based On NetHogs

With the rapid development of Internet, the growing popularity of network applications and increasingly rapid maturation of cloud computing technology, In the world, pc, mobile phone, pad of the various app on people’s lives and work has brought great convenience. Various O2 O, various app greatly enriched the various needs of our material life, fast bring people great experience. At the same time the rapid development of the Internet, business needs and data traffic in the mobile terminal and the pc side is also growing. Occupy a small amount of bandwidth from the early web browsing,sending and receiving e-mail, QQ chat at this stage to take up a lot of bandwidth Thunder video, Youku video, a variety of network games and so on. It can be seen with the development of mobile multimedia, this pc and mobile terminal app also increasingly abundant. But also bandwidth network spam messages, such as advertising,video and other garbage. Besides Internet content security has become increasingly serious. How effective detection and identification of network traffic packets,large-scale network traffic identification technology need to study the issue. Currently,there are already a number of large-scale clustering traffic data monitoring program. But they are more of a client-side detection pc, for now there is no large-scale clustered server-side statistics and analyzing traffic data packets. How to design a detector clustered servers to provide server-side application-layer inspection capabilities to meet the limit or cut off network traffic filtering, Traffic recognition features such as the cluster server is the main topic of this thesis.In this thesis, the flow identification technology and flow control cloud environments. Detailed design cluster server architecture to detect, from the inside of stratus cloud computing statistics server application server and an outer layer composed of such a design can use the ability of large-scale computing in the clouds, and can detect and count the outer cloud application server data traffic. Detailed study of open source linux packet capture tool Net Hogs. Hierarchical design flow identification scheme on the basis of Net Hogs, is added massive traffic identification algorithm.Thereby greatly improves the efficiency and accuracy of the outer layer of the cloud detection packet. And the detailed design and implementation of this traffic system modules, mainly by the flow control module, traffic management module, the databasemodule, client module flow detection, flow detection server module. Finally, while the realization of traffic detection system after in-depth testing and analysis.Significance of this thesis lies in cloud computing and parallel computing means of the idea, proposed architecture suitable for monitoring traffic monitoring clustered servers. And based on this architecture of the traffic carried on the detailed design and implementation. While the realization of traffic detection systems deployed to the lab.Detection accuracy and efficiency of the flow detection system in practical application.And the final test results are analyzed in detail the features and performance. Detailed summary of the results of the analysis of data obtained, derived shortcomings of the present system and the flow detection areas for improvement.