Dynamics of heterogeneous peer-to-peer networks

Fernando Paganini, Andrés Ferragut, Martín Zubeldía

Research output: Chapter in Book/Report/Conference proceedingConference contribution

2 Scopus citations

Abstract

The most tractable models of population dynamics in peer-to-peer file sharing systems apply to a single class of peers with homogeneous network access parameters. When upload bandwidths are heterogeneous, reciprocity mechanisms lead to non-uniform download rates and a more complex multi-class dynamics. We consider first a model where mutual download bandwidths are allocated in proportion to the upload speed, plus a uniformly distributed server component. For an ordinary differential equation model of the multi-class peer populations, we characterize the equilibrium and establish its global stability, invoking results from monotone systems. We also analyze a partial differential equation model that tracks download progress of the populations; we establish the local asymptotic stability of the equilibrium. Finally, we extend the ODE model to include a mix of proportional and uniform bandwidth allocation, which better describes the mechanisms of BitTorrent systems; again we characterize equilibrium configurations and give a partial result on local stability.

Original languageEnglish (US)
Title of host publication2013 IEEE 52nd Annual Conference on Decision and Control, CDC 2013
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages3293-3298
Number of pages6
ISBN (Print)9781467357173
DOIs
StatePublished - 2013
Externally publishedYes
Event52nd IEEE Conference on Decision and Control, CDC 2013 - Florence, Italy
Duration: Dec 10 2013Dec 13 2013

Publication series

NameProceedings of the IEEE Conference on Decision and Control
ISSN (Print)0743-1546
ISSN (Electronic)2576-2370

Other

Other52nd IEEE Conference on Decision and Control, CDC 2013
Country/TerritoryItaly
CityFlorence
Period12/10/1312/13/13

Fingerprint

Dive into the research topics of 'Dynamics of heterogeneous peer-to-peer networks'. Together they form a unique fingerprint.

Cite this