Robust network dimensioning for realtime services over IP networks with traffic deviation

Abstract

Network planning for realtime services in traditional systems is normally performed based on a fixed traffic demand. Realtime services such as voice and video are currently being widely deployed over packet-based IP networks. Due to the lack of resource control mechanisms in IP networks, the actual traffic during network operation may easily deviate from the given mean traffic matrix values. As realtime services impose stringent quality of service (QoS) constraints, any traffic deviation inside the network might lead to severe quality degradation if any of the links provides insufficient service rate for the available traffic at any time. In this work, we highlight the significance of this problem and propose an effective approach to provide a highly robust network against traffic deviation. By means of both analytical and simulation studies, we study the distribution of realtime connections over an enterprise IP network and show how capacity needs of realtime traffic are affected when various resource control mechanisms are employed. In addition, we study the resulting tradeoff between capacity needs and traffic QoS parameters including packet loss and blocking probability.