Collaboration between ISPs for Efficient Overlay Traffic Management - Inria - Institut national de recherche en sciences et technologies du numérique Accéder directement au contenu
Communication Dans Un Congrès Année : 2011

Collaboration between ISPs for Efficient Overlay Traffic Management

Résumé

As peer-to-peer (P2P) applications (e.g. BitTorrent) impose high costs to Internet Service Providers (ISPs) due to the large volumes of inter-domain traffic generated, extensive research work has been done concerning locality-awareness approaches. Although such approaches are based on properties of the physical network topology, they take very little consideration of the real business relationships between ISPs (peering and transit agreements) as well as of the heterogeneous peer distributions among ISPs. In this paper, we propose an innovative way to exploit the business relationships between ISPs of either the same or different Tiers, by introducing new collaborative approaches for overlay traffic management on top of the locality-awareness ones. By means of simulations, we show that win-win situation (reduced transit traffic for ISPs and better performance for users) can mostly be achieved under certain complicated collaborative approaches. Also we show how the collaboration between the upper Tier ISPs can affect favorably both the inter-domain traffic in the transit links, where charging applies, and the performance of their customers ISPs and how the latter can benefit from their collaboration with the transit ISPs.
Fichier principal
Vignette du fichier
978-3-642-20798-3_9_Chapter.pdf (344.75 Ko) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)
Loading...

Dates et versions

hal-01597981 , version 1 (29-09-2017)

Licence

Paternité

Identifiants

Citer

Eleni Agiatzidou, George D. Stamoulis. Collaboration between ISPs for Efficient Overlay Traffic Management. 10th IFIP Networking Conference (NETWORKING), May 2011, Valencia, Spain. pp.109-120, ⟨10.1007/978-3-642-20798-3_9⟩. ⟨hal-01597981⟩
50 Consultations
56 Téléchargements

Altmetric

Partager

Gmail Facebook X LinkedIn More