A simple proactive provider participation technique in a mesh-based peer-to-peer streaming service

Authors:
Darío Padula;María Elisa Bertinat;Franco Robledo Amoza;Pablo Rodríguez-Bocca;Pablo Romero
Affiliations:
Laboratorio de Probabilidad y Estadística, Facultad de Ingeniería, Universidad de la República, Montevideo, Uruguay;Laboratorio de Probabilidad y Estadística, Facultad de Ingeniería, Universidad de la República, Montevideo, Uruguay;Laboratorio de Probabilidad y Estadística, Facultad de Ingeniería, Universidad de la República, Montevideo, Uruguay;Laboratorio de Probabilidad y Estadística, Facultad de Ingeniería, Universidad de la República, Montevideo, Uruguay;Laboratorio de Probabilidad y Estadística, Facultad de Ingeniería, Universidad de la República, Montevideo, Uruguay
Venue:
HAIS'11 Proceedings of the 6th international conference on Hybrid artificial intelligent systems - Volume Part II
Year:
2011

Citing 3
Cited 0

P4p: provider portal for applications

Proceedings of the ACM SIGCOMM 2008 conference on Data communication
GoalBit: the first free and open source peer-to-peer streaming network

Proceedings of the 5th International Latin American Networking Conference
A trust model in P4P-integrated P2P networks based on domain management

ISCC '10 Proceedings of the The IEEE symposium on Computers and Communications

Quantified Score

Hi-index	0.00

Visualization

Abstract

The design of efficient protocols for mesh-based Peer-to-Peer (P2P) networks has many challenges, one of them is bandwidth allocation. On one hand, peers demand high Quality of Experience and network traffic when they watch their streaming contents. On the other, Internet Service Providers (ISPs) support their business with the capacity of their international links. A recent strategy considered in order to meet both requirements is the Proactive Provider Participation, shortly named P4P [1], [2]. This paper addresses the multi-objective P4P problem for live video content. First, we introduce a Ford-Fulkerson-based solution, that solves the P4P problem when only one content is distributed. Second, a greedy randomized technique is applied when multiple contents are shared. Finally, our algorithm is tested in a real multiple-content scenario, showing that the results highly outperform current strategies.