DocumentCode :
1686784
Title :
Fluid Limits Applied to Peer to Peer Network Analysis
Author :
Aspirot, Laura ; Mordecki, Ernesto ; Rubino, Gerardo
Author_Institution :
Univ. de la Republica, Montevideo, Uruguay
fYear :
2011
Firstpage :
13
Lastpage :
20
Abstract :
The objective of several techniques including fluid limits and mean field approximations is to analyze a stochastic complex system (e.g. Markovian) studying a simplified model (deterministic, represented by ordinary differential equations (ODEs)). In this paper, we explore models proposed for the analysis of BitTorrent P2P systems and we provide the arguments to justify the passage from the stochastic process, under adequate scaling, to a fluid approximation driven by an ODE. We also make the link between the stationary regime of the stochastic models and the fixed points of the associated ODEs. Finally, we analyze the asymptotic distribution of the scaled process.
Keywords :
Markov processes; approximation theory; differential equations; peer-to-peer computing; BitTorrent P2P systems; Markovian system; fluid approximation; fluid limits; mean field approximations; ordinary differential equations; peer to peer network analysis; stochastic complex system; Analytical models; Approximation methods; Convergence; Markov processes; Mathematical model; Peer to peer computing; BitTorrent; fluid limits; mean fields;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Quantitative Evaluation of Systems (QEST), 2011 Eighth International Conference on
Conference_Location :
Aachen
Print_ISBN :
978-1-4577-0973-9
Type :
conf
DOI :
10.1109/QEST.2011.11
Filename :
6042025
Link To Document :
بازگشت