A Content Prepositioning Using Popularity Prediction in Hybrid Peer-to-Peer Network with Cloud Storage

Abstract

This paper proposes a digital contents prepositioning method based on the prediction of the popularity trend for content sharing in an environment where Peer-to-Peer (P2P) network and cloud storage service are used together. In the proposed method, the popularity trend of contents is predicted by linear approximation, the contents are prepositioned over the P2P network, and placed in the cloud storage if necessary. The usefulness of the proposed method was verified using computer simulations from the viewpoints of network load, content loss, and cloud storage usage cost, changing the interval of preliminary content replica redeployment and the peers' storage capacity.