Optimal bandwidth sharing in multiswarm multiparty P2P video-conferencing systems

  • Authors:

  • Chao Liang;Miao Zhao;Yong Liu

  • Affiliations:

  • Department of Electrical and Computer Engineering, Polytechnic Institute of New York University, Brooklyn, NY;Department of Electrical and Computer Engineering, State University of New York at Stony Brook, Stony Brook, NY;Department of Electrical and Computer Engineering, Polytechnic Institute of New York University, Brooklyn, NY

  • Venue:
  • IEEE/ACM Transactions on Networking (TON)
  • Year:
  • 2011

Quantified Score

Hi-index 0.00

Visualization

Abstract

In a multiparty video conference, multiple users simultaneously distribute video streams to their receivers. As the traditional server-based solutions incur high infrastructure and bandwidth cost, conventional peer-to-peer (P2P) solutions only leveraging end-users' upload bandwidth are normally not self-sustainable: The video streaming workload increases quadratically with the number of users as each user could generate and distribute video streams, while the user upload bandwidth only increases linearly. Recently, hybrid solutions have been proposed that employ helpers to address the bandwidth deficiency in P2P video-conferencing swarms. It is also noticed that a system hosting multiple parallel conferencing swarms can benefit from cross-swarm bandwidth sharing. However, how to optimally share bandwidth in such systems has not been explored so far. In this paper, we study the optimal bandwidth sharing in multiswarm multiparty P2P video-conferencing systems with helpers and investigate two cross-swarm bandwidth-sharing scenarios: 1) swarms are independent and peers from different swarms share a common pool of helpers; 2) swarms are cooperative and peers in a bandwidth-rich swarm can further share their bandwidth with peers in a bandwidth-poor swarm. For each scenario, we develop distributed algorithms for intraswarm and interswarm bandwidth allocation under a utility-maximization framework. Through analysis and simulation, we show that the proposed algorithms are robust to peer dynamics and can adaptively allocate peer and helper bandwidth across swarms so as to achieve the system-wide optimum.