Overload control in QoS-aware web servers

With the explosive use of Internet, contemporary web servers are susceptible to overloads during which their services deteriorate drastically and often lead to denial of services. Overloads are of more serious concerns for QoS-aware servers. Evaluation of performance of QoS-aware servers in terms of the number of request completion is not very meaningful. A better measure would be the number of completed sessions. In this paper, we proposed two methods to prevent and control overloads in web servers by utilizing session-based relationship among HTTP requests. We first exploited the dependence among session-based requests by analyzing and predicting the reference patterns. Using the dependency relationships, we have derived traffic conformation functions that can be used for capacity planning and overload prevention in web servers. Second, we have proposed a dynamic weighted fair sharing (DWFS) scheduling algorithm to control overloads in web servers. DWFS is distinguished from other scheduling algorithms in the sense that it aims to avoid processing of requests that belong to sessions that are likely to be aborted in the near future. The experimental results demonstrate that DWFS can improve server responsiveness by as high as 50% while providing QoS support through service differentiation for a class of application environment. (C) 2003 Elsevier Science B.V. All rights reserved.