A Measurement and Modeling Study of P2P IPTV Applications

P2P IPTV applications start to be very popular on the Internet. Previous work on measuring and analyzing P2P IPTV traffic mostly focuses on a specific type of P2P applications or the behavior characteristics of P2P IPTV nodes acting as a client. In this paper, we have undertaken an in-depth measurement and modeling study of the most common P2P IPTV applications, specially focusing on the behavior characteristics of peers acting as a server. We measure and model the distributions of flow inter-arrival time, arrival rate, duration and size. Our results show the following. The statistics can fit equally well into distributions. Due to the different strategies, there exhibit some differences in the traffic characteristics. But, because of P2P IPTV users behaving similarly, they mainly have similar distributions and the traffic of them all have heavy-tailed phenomenon. Insights obtained in this study will be valuable for the identification of P2P IPTV traffic.