Performance Bounds for a Cascade based Multifractal Traffic Model with Generalized Multiplier Distributions

Abstract

Abstract-In this paper we propose a multifractal traffic model that is based on a multiplicative cascade with generalized multiplier distributions (CGMD). The multipliers are determined through their probability densities estimated from real network traffic flows by using Kernel and Acceptance/Rejection methods. Statistical analysis and queueing behavior studies were carried out for the model validation in comparison to other multiplicative cascade based models. In order to build an efficient estimation method of performance bounds for network traffic flows that takes account of multifractal characteristics, we derive the effective bandwidth for the CGMD model as well as its Hurst parameter. The proposed performance bounds are computed by relating the CGMD based effective bandwidth to statistical network calculus concepts. Our performance bound estimation approach is evaluated through simulations with Internet and Ethernet traffic traces, verifying its efficiency in describing the byte loss probability and mean buffer occupation.

Adaptive Rate Control Based on Loss Probability Estimation Considering a Cascade Based Multifractal Model for the Network Traffic

“…Among the proposals for rate control and resource allocation, we highlight the adaptive algorithm presented in Ref. 31, which considers monofractal characteristics of network traffic. We describe this adaptive algorithm in Sec.…”

“…31, the authors consider the DRA as an efficient method for dynamic rate allocation when compared to the fixed service rate, since the adaptive scheme has better resource utilization while providing the same Quality of Service (QoS) guarantee. However, the DRA may be unable to efficiently capture network traffic characteristics, such as persistent bursts of traffic in different time scales.…”

“…10,23,25 Some studies have revealed that multifractal models are suitable to describe different characteristics of network traffic. 10,23,25,30,31 In fact, analysis and tests were carried out for various network traffic traces, showing that the multifractal model is suitable for video traffic, 9 LAN traffic (Local Area Network), 16,23 WAN (Wide Area Network) 10,24 and World Wide Web (WWW) traffic type. 18 The more accurate the modeling of the network flow behavior, the more appropriate the service can be offered to the user.…”