Server Guaranteed Cap: An Incentive Mechanism for Maximizing Streaming Quality in Heterogeneous Overlays

Abstract

We address the problem of maximizing the social welfare in a peer-to-peer streaming overlay given a fixed amount of server upload capacity. We show that peers' selfish behavior leads to an equilibrium that is suboptimal in terms of social welfare, because selfish peers are interested in forming clusters and exchanging data among themselves. In order to increase the social welfare we propose a novel incentive mechanism, Server Guaranteed Cap (SGC), that uses the server capacity as an incentive for high contributing peers to upload to low contributing ones. We prove that SGC is individually rational and incentive compatible. We also show that under very general conditions, there exists exactly one server capacity allocation that maximizes the social welfare under SGC, hence simple gradient based method can be used to find the optimal allocation.

Using topology information for quality‐aware Peer‐to‐Peer multilayer video streaming

“…An unbalanced data exchange is a problem that decreases a P2P system's potential performance in terms of bandwidth utilization [13][14][15][16]. Both the unwillingness to share by users and inferior data exchanges as a result of the distributed algorithms form a huge challenge when designing a P2P streaming applications.…”

“…Constraint (14) ensures that when a destination node d receives video layer l i+1 , also, video layer l i is received on one of d's direct incoming links. Constraint (15) imposes that when a destination node d receives layer l, there must be (at least) a virtual path starting from the injector node z.…”

Using Centrality Metrics to Predict Peer Cooperation in Live Streaming Applications

“…Silverston et al [17] proposed a different incentive mechanism where peers unable to serve data should answer with pointers to other peers that are able to serve. Chatzidrossos et al [4] argued that many noncooperative peers are unable to upload as much data as they download due asymmetric DSL or HSDPA connections. To maximize the social welfare, they proposed to reserve a fraction of server upload bandwidth to altruistic peers that upload more data than they download.…”

“…There are various techniques for detecting non-cooperative peers [2,8,14] and incentivizing peer cooperation [4,16,17] in live streaming applications. Our motivation for investigating the benefit of exploiting centrality metrics for that purposes lies in the observation that many P2P live applications keep a centralized server (tracker) that periodically receives control messages from each peer.…”

“…Moreover, some techniques collect the ratio of upload to download periodically from peers and use authentication mechanisms to be robust to malicious peers, which increases communication and computation overhead in the system. We believe that our strategy might be jointly applied with such techniques by complementing them and reducing their overhead [4,16].…”