Abstract
Video-on-demand (VoD) streaming based on Peer-to-Peer (P2P) overlay has become one of the most popular applications over the Internet. However, the bottlenecks in the available upload bandwidth, both at the media source and inside the overlay, is still a challenging problem, which may limit the satisfaction of users in watching video. A potential solution for this problem is assisting the P2P streaming overlay by a cloud computing infrastructure to guarantee the minimum level of users' satisfaction: rented cloud resources (p rs) are added on demand to the overlay, to increase the amount of available bandwidth and the probability of receiving the video streaming on time. In such an approach, the VoD service provider relies on helpers to deliver video chunks to users and needs to pay for the cloud bandwidth consumption. Hence, under the limitation on cloud bandwidth consumption, how a helper allocates its bandwidth resource for chunk delivering is important to improve the satisfaction degrees of users. In this paper, we present an analytical optimization model to tackle with the bandwidth allocation problem for helpers. Furthermore, a greedy-based policy aiming to improve user satisfaction is proposed based on the model. We perform extensive simulations and the experimental results show that compared to existing policies, the proposed policy can provide high satisfaction degrees of users.