The enormous growth in popularity of peer-to-peer applications has recently introduced great interest in understanding the associated traffic workload and behavior. The goal of this work is determining the fundamental dynamics characterizing such traffic that can be used to develop simple and effective prediction models and to illustrate and describe fundamental performance issues. The discovery of nonlinear traffic dynamics, due to the very complex characteristics of the involved time series, led us to use several nonlinear analysis techniques and tools evidencing the presence of chaos-related structures together with selfsimilarity and long-range dependence features. © 2010 IEEE.
Insights into peer to peer traffic through nonlinear analysis
Fiore, Ugo
2010-01-01
Abstract
The enormous growth in popularity of peer-to-peer applications has recently introduced great interest in understanding the associated traffic workload and behavior. The goal of this work is determining the fundamental dynamics characterizing such traffic that can be used to develop simple and effective prediction models and to illustrate and describe fundamental performance issues. The discovery of nonlinear traffic dynamics, due to the very complex characteristics of the involved time series, led us to use several nonlinear analysis techniques and tools evidencing the presence of chaos-related structures together with selfsimilarity and long-range dependence features. © 2010 IEEE.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
