Traffic Modeling and Performance Analysis of Commercial Web Sites Cathy H. Xia, Zhen Liu, Mark S. Squillante, Li Zhang, Naceur Malouch IBM Research Division Thomas J. Watson Research Center Yorktown Heights, NY 10598 {cathyx, zhenl, ross, zhangl} @us.ibm.com Naceur.Malouch@ sophia.inri~fr With the rapid advances in Internet technology, e-commerce is becoming a mature business strategy. The concept of Quality of Service (QoS) is working its way to the front lines of e-business commitments and requirements as it plays an important role in Internet applications, services and pricing negotiations. One needs to have a fundamental understanding of the key characteristics of the workload patterns in commercial Web sites and a fundamental understanding of the impact of such workload patterns on Web server performance as well as the server capacity required to guarantee a certain level of QoS, e.g., user request response time. Our primary focus in this study is to investigate these important research issues related to such performance metrics in the context of a real commercial Web server environment. A key aspect of our study must therefore concern the arrival patterns of incoming user requests and the resource requirements for serving these requests, as these represent two of the
/lp/association-for-computing-machinery/traffic-modeling-and-performance-analysis-of-commercial-web-sites-t4SHKVhvCU