Optimal Peer Selection Strategy in P2P-VoD Systems Using Dynamic Evolution Strategy

Abstract

Peer-to-Peer Video-on-Demand (VoD) systems are rising as a new dominant way to distribute video content over IP networks. Although this approach improves the overall VoD system scalability, it still poses new challenges such as peers resource allocation. There has been numerous research works versed into the P2P streaming with different focus areas, and different approaches. Some work cover the resource allocation issue in P2P streaming systems where the real-time streaming add another dimension to the problem. Most work on P2P resource allocation approaches the problem with static rules strategies that fail to dynamically adjust in face of changing content demand (popularity) trends. In this paper, we focus on the problem of enhancing the performances of a P2P system by adapting the peer allocation strategy. The proposed resource allocation system dynamically switches between multiple strategies to optimally respond to observed and predicted shifts in the content popularity. To do so, a dynamic estimation problem is solved using a Levy distribution based Dynamic Evolution Strategy algorithm. The obtained results show that using a dynamic resource allocation reduces the rejection rate while maintaining high diversification in the face of a dynamically changing title demand.