Minimizing weighted sum download time for one-to-many file transfer in peer-to-peer networks

TL;DR

This paper addresses minimizing weighted sum download time in P2P networks with heterogeneous bandwidth constraints, proposing static and dynamic schemes that improve efficiency and are backed by convex optimization and empirical results.

Contribution

It introduces a polynomial-time optimization for static scenarios with heterogeneous peers and proposes a dynamic rateless-coding scheme that outperforms static methods.

Findings

Static optimization achieves near-optimal WSDT.

A tight lower bound for minimum WSDT is established.

Dynamic rateless-coding significantly reduces WSDT compared to static schemes.

Abstract

This paper considers the problem of transferring a file from one source node to multiple receivers in a peer-to-peer (P2P) network. The objective is to minimize the weighted sum download time (WSDT) for the one-to-many file transfer. Previous work has shown that, given an order at which the receivers finish downloading, the minimum WSD can be solved in polynomial time by convex optimization, and can be achieved by linear network coding, assuming that node uplinks are the only bottleneck in the network. This paper, however, considers heterogeneous peers with both uplink and downlink bandwidth constraints specified. The static scenario is a file-transfer scheme in which the network resource allocation remains static until all receivers finish downloading. This paper first shows that the static scenario may be optimized in polynomial time by convex optimization, and the associated optimal static WSD can be achieved by linear network coding. This paper then presented a lower bound to the minimum WSDT that is easily computed and turns out to be tight across a wide range of parameterizations of the problem. This paper also proposes a static routing-based scheme and a static rateless-coding-based scheme which have almost-optimal empirical performances. The dynamic scenario is a file-transfer scheme which can re-allocate the network resource during the file transfer. This paper proposes a dynamic rateless-coding-based scheme, which provides significantly smaller WSDT than the optimal static scenario does.