Minimizing Weighted Sum Finish Time for One-to-Many File Transfer in Peer-to-Peer Networks

Abstract—This paper considers the problem of transfer-ring a file from one source node to multiple receivers in a peer-to-peer (P2P) network. The objective is to minimize the weighted sum finish time (WSFT) for the one-to-many file transfer where peers have both uplink and downlink bandwidth constraints specified. The static scenario is a file-transfer scheme in which the constructed network topology and the network resource (link throughput) allo-cation remains static until all receivers finish downloading. This paper first shows that the static scenario can be op-timized in polynomial time by convex optimization, and the associated optimal static WSFT can be achieved by linear network coding. This paper also proposes a static rateless-coding-based scheme which has almost-optimal empirical performance. The dynamic scenario is a file-transfer scheme which can re-construct the network topol-ogy and re-allocate the network resource during the file transfer. This paper proposes a dynamic rateless-coding-based scheme, which provides significantly smaller WSFT than the optimal static scheme does. Index Terms — P2P network, network coding, rateless code, static scenario, dynamic scenario