Modeling and analysis of a p2p-vod system based on stochastic network calculus

  • Authors:
  • Kai Wang;Yuming Jiang;Chuang Lin

  • Affiliations:
  • Department of Computer Science and Technology, Tsinghua University, China and Department of Computing and Mathematical Sciences, Caltifornia Institute of Technology;Department of Telematics & Q2S Center, Norwegian University of Science and Technology, Norway;Department of Computer Science and Technology, Tsinghua University, China

  • Venue:
  • MMB'12/DFT'12 Proceedings of the 16th international GI/ITG conference on Measurement, Modelling, and Evaluation of Computing Systems and Dependability and Fault Tolerance
  • Year:
  • 2012

Quantified Score

Hi-index 0.00

Visualization

Abstract

Peer-to-peer video-on-demand (P2P-VoD) is one of the most commercially important and technically challenging applications. In such systems, each video file is typically divided into sub-pieces that are stored possibly at different sending peers, and the receiving peer often needs to coordinate multiple senders by specifying a transmission schedule for them. In this paper, we investigate this transmission scheduling problem for a specific P2P-VoD system which is PPLive. We formalize the problem of scheduling sub-piece transmission based on stochastic network calculus, and analyze its delay performance. Based on the delay analysis, an optimization problem is formulated to design the schedule for sending peers to transmit their sub-pieces. The objective of this optimization is to maximize a defined reward and at the same time meet the delay constraint required to ensure video quality at the receiver.