Resource Assignment in Box-Based P2P Video-on-Demand Systems

Abstract—Video-on-demand (VoD) systems that make use of the storage capacities at set-top boxes to assist the streaming servers have been introduced recently. In these systems, videos are separated into substreams and randomly stored in boxes, which are organized in a P2P network. By this way, this architecture combines the load balancing and fault tolerating features of P2P systems with the stability of set-top boxes, since they usually have much longer online time than traditional PC based peers. The VoD service provider controls two different re-sources: the allocation of substreams to the selected boxes, and the parameters associated to the streaming servers (number, bandwidth, storage, placement in the network). In this work, we present the Resource Assignment Problem (RAP) which tries to optimize the previous resources in order to reduce the associated costs. This problem is presented as a linear programming problem and it is solved using the MATLAB optimization toolbox. We have evaluated the influence of the bandwidth and the placement of the server in the Internet. Keywords-VoD; P2P; set-top box; linear programming. I