Web servers face all kinds of users; some of them engage malicious activities to degrade or completely block network services, such as flooding attacks. As a result, lots of resource and bandwidth on web sites might be wasted. In this paper, we propose a detection method based on reverse proxy servers to detect the flooding attacks. Our proposed method can be incorporated in a priority queue-based scheme to enhance quality of service for normal users. We have done performance evaluation of our method by implementing it on a server under heavy load. The experimental results show that our method is effective to detect malicious flooding attacks and is efficient to analyze and classify network traffic. Since the process time of the detection method grows linearly with the data size, it is permissible to implement our proposed scheme with servers under heavy load to improve their performance. ?2008 IEEE.
Relation:
Proceedings - 8th International Conference on Intelligent Systems Design and Applications, ISDA 2008 3 , art. no. 4696475 , pp. 281-284
