Towards Green P2P: Understanding the Energy Consumption in P2P under Content Pollution
Peiqing Zhang, Bjarne E. Helvik · 2010
Peer-to-peer distribution of information plays an increasingly important role in nowadays society. Currently, the major fraction of the Internet traffic stem from peer-to-peer systems. Hence, in order to cut down the emissions of carbon dioxide from ICT (information and communication technology) business, which contributes 2 percent of the global consumption, it is necessary to understand energy consumption in peer-to-peer systems. In this paper, a model of this is developed, taking into consideration content pollution, the most common attack in peer-to-peer, as well as parameters describing the user behavior like peer churn, pollution detecting rate and the connectivity requirements. The model is analyzed analytically and numerically to reveal how these factors influence the overall power consumption in both steady state and flash crowd information exchange scenarios. An optimal, from the energy efficiency perspective, peer online holding time of clean copies is found which is dependent on other system parameters.