Quality of Service Based Pricing Schemes for Content Sharing in Peer-to-Peer Networks

Abstract

In this paper, we study quality-of-service (QoS) based pricing schemes that serve as incentive mechanisms to induce sharing behaviors in Peer-to-Peer (P2P) networks. We incorporate operational QoS metrics into users' utility functions and demonstrate how they affect individual users' content sharing decisions. Using a game-theoretic model, our study reveals how organizations respond to the changes of operational QoS metrics in their design of pricing schemes for various business objectives at different stages of network evolution. Our results show that a higher upload capacity can foster rational sharing to start when the network is small; however, it also discourages sharing behaviors when the network becomes large. In order to induce a socially optimal behavior, a pricing scheme will not charge users for requesting content while compensating them for sharing content. Such compensation is found to increase faster with the network size when the network is large. In order to maximize the profit of a monopolistic provider, however, a pricing scheme will charge content requests with a positive price while providing less compensation to sharing users compared to the socially optimal scheme. When the network size is small, such compensation can be even negative, which implies that a monopolistic provider discourages content sharing when the network is small, but encourages it when the network becomes larger. In addition, we find that more information about peer upload capacity discourages peers to share.