A Performance Evaluation Study of Video-on-Demand Traffic over IP Networks

Low cost, high bandwidth Internet connectivity has made Video-on-Demand (VoD) one of the most significant new age technologies being adopted by ever-growing users. Video applications such as online video streaming, video conferencing and many others are being increasingly used. The increased in IP based video applications has resulted in more traffic over IP networks. Hence, there is a need for efficient network bandwidth utilization for VoD applications by employing VoD systems which are more efficient and better in performance. Various architectures have been proposed and implemented to handle the rapid growth in demand for video delivering technologies. In this paper we have thoroughly examined some of the proposed VoD architectures using simulation tool. The simulation tool that we used provided an efficient platform to analyze different architectures and obtain performance metrics and plots. The objective of this paper is to analyze the VoD traffic, in particular from proxy and VoD servers. Hence the primary focus is on the Proxy and CDN architectures, which deploy proxy servers at the edge of the network. We have statistically analyzed the VoD traffic through experiments, under different network conditions and have derived suggestions for optimal network parameters required for efficient VoD architectures. Carefully designed proxy-assisted schemes can improve the performance of VoD network, thereby making VoD systems more reliable and scalable.