On the higher moments of TCP

Abstract In this paper we describe the moments of a stochastic model of the Additive Increase Multiplicative Decrease (AIMD) algorithm. AIMD is the algorithm that underpins the Transmission Control Protocol (TCP), which is used extensively in the internet. We prove that the Markov chain describing TCP has the remarkable property that all moments converge to their asymptotes at exactly the same rate. Further, we illustrate how a closed form solution can be obtained from the network properties, and this formula is explicitly calculated for the case of the third moment.

[1]  R. Srikant,et al.  TCP-Illinois: A loss- and delay-based congestion control algorithm for high-speed networks , 2008, Perform. Evaluation.

[2]  Robert Shorten,et al.  On the second eigenvalue of matrices associated with TCP , 2006 .

[3]  James A. Yorke,et al.  Stability of TCP Dynamics in Large Data Networks , 2009, SIAM J. Appl. Dyn. Syst..

[4]  Charles R. Johnson,et al.  Matrix analysis , 1985, Statistical Inference for Engineers and Data Scientists.

[5]  François Baccelli,et al.  AIMD, fairness and fractal scaling of TCP traffic , 2002, Proceedings.Twenty-First Annual Joint Conference of the IEEE Computer and Communications Societies.

[6]  Robert Shorten,et al.  On the Dynamics of TCP's Higher Moments , 2007, IEEE Communications Letters.

[7]  P. Lancaster,et al.  The theory of matrices : with applications , 1985 .

[8]  F. Baccelli,et al.  Interaction of TCP flows as billiards , 2003, IEEE INFOCOM 2003. Twenty-second Annual Joint Conference of the IEEE Computer and Communications Societies (IEEE Cat. No.03CH37428).

[9]  Fabian R. Wirth,et al.  A positive systems model of TCP-like congestion control: asymptotic results , 2006, IEEE/ACM Transactions on Networking.

[10]  F. R. Gantmakher.  The Theory of Matrices , 1984 .