Stratification in P2P Networks: Application to BitTorrent

  • Authors:
  • Anh-Tuan Gai;Fabien Mathieu;Fabien de Montgolfier;Julien Reynier

  • Affiliations:
  • INRIA, Cedex, France;Orange Labs, France;LIAFA - Université Paris 7, France;Motorola Labs, France

  • Venue:
  • ICDCS '07 Proceedings of the 27th International Conference on Distributed Computing Systems
  • Year:
  • 2007

Quantified Score

Hi-index 0.00

Visualization

Abstract

We consider a model for decentralized collaborative networks that is based on stable matching theory. This model is applied to systems with a global ranking utility function, which admits a unique stable configuration. We study the speed of convergence and analyze the stratification properties of the stable configuration, both when all collaborations are possible and for random possible collaborations. As a practical example, we consider the BitTorrent Titfor- Tat policy. For this system, our model provides an interesting insight into peer download rates and a good understanding of upload settings strategy.