Enrich multi-channel P2P VoD streaming based on dynamic replication strategy

Abstract

Peer-to-Peer Video-on-Demand (VoD) is a favorable solution which compromises thousands of videos to millions of users with completeinteractive video watching stream. Most of the profitable P2P streaming groupsPPLive, PPStream and UUSee have announced a multi-channel P2P VoD system that approvals user to view extra one channel at a time. The present multiple channel P2P VoD system resonant a video at a low streaming rate due to the channel resource inequity and channel churn. In order to growth the streaming capacity, this paper highlights completely different effective helpers created resource balancing scheme that actively recognizes the supply-and-demand inequity in multiple channels. Moreover, peers in an extra channel help its unused bandwidth resources to peers in a shortage channel that minimizes the server bandwidth consumption. To provide a desired replication ratio for optimal caching, it develops a dynamic replication strategy that optimally tunes the number of replicas based on dynamic popularity in a distributed and dynamic routine. This work accurately forecasts the varying popularity over time using Auto-Regressive Integrated Moving Average (ARIMA) model, an effective time-series forecasting technique that supports dynamic environment. Experimental assessment displays that the offered dynamic replication strategy which should achieves high streaming capacity under reduced server workload when associated to existing replication algorithms.