Dynamical Modelling of TCP Packet Traffic on Scale-Free Networks

The interactive growth method is used to model the topology of real networks. Packet traffic is simulated crossing this network using the closed-loop packet transfer mechanism Transmission Control Protocol. Comparisons are made for traffic on regular and scale-free networks with open-loop and closed-loop packet transfer mechanisms. Packet lifetimes and queue behaviour for long range dependent sources (LRD) are compared with short range dependent Poisson sources (SRD) at the same loadings. The effects of varying server strengths are studied as are the results of imposing packet loss. The robustness of results is tested by varying patterns of hosts and using different networks with similar parameters. A marked difference is seen between outputs from the two source types, SRD and LRD, emphasizing that long range dependence in sources is an important factor. Changing host patterns for interactive growth networks produces very similar results indicating a good degree of robustness in the simulations. However, these results are very different from those obtained for regular and scale-free network simulations using an open-loop transfer mechanism. This demonstrates the need for more accurate models such as the interactive growth model, and for the simulation of closed-loop algorithms such as transmission control protocol.

[1]  Walter Willinger,et al.  Self-similar traffic and network dynamics , 2002, Proc. IEEE.

[2]  Albert-László Barabási,et al.  Statistical mechanics of complex networks , 2001, ArXiv.

[3]  Xiao Fan Wang,et al.  Complex Networks: Topology, Dynamics and Synchronization , 2002, Int. J. Bifurc. Chaos.

[4]  Shi Zhou,et al.  Redundancy and Robustness of the AS-level Internet topology and its models , 2004, ArXiv.

[5]  Alessandro Vespignani,et al.  Large-scale topological and dynamical properties of the Internet. , 2001, Physical review. E, Statistical, nonlinear, and soft matter physics.

[6]  Jonathan M. Pitts,et al.  Internet packet traffic congestion , 2003, Proceedings of the 2003 International Symposium on Circuits and Systems, 2003. ISCAS '03..

[7]  Walter Willinger,et al.  The origin of power laws in Internet topologies revisited , 2002, Proceedings.Twenty-First Annual Joint Conference of the IEEE Computer and Communications Societies.

[8]  Michalis Faloutsos,et al.  On power-law relationships of the Internet topology , 1999, SIGCOMM '99.

[9]  Shi Zhou,et al.  Towards Modelling The Internet Topology - The Interactive Growth Model , 2003, ArXiv.

[10]  Murad S. Taqqu,et al.  On the Self-Similar Nature of Ethernet Traffic , 1993, SIGCOMM.

[11]  H. Hurst.  METHODS OF USING LONG-TERM STORAGE IN RESERVOIRS. , 1956 .

[13]  M Woolf,et al.  Optimization and phase transitions in a chaotic model of data traffic. , 2002, Physical review. E, Statistical, nonlinear, and soft matter physics.

[14]  Randy H. Katz,et al.  Characterizing the Internet hierarchy from multiple vantage points , 2002, Proceedings.Twenty-First Annual Joint Conference of the IEEE Computer and Communications Societies.

[15]  Donald F. Towsley,et al.  On distinguishing between Internet power law topology generators , 2002, Proceedings.Twenty-First Annual Joint Conference of the IEEE Computer and Communications Societies.

[16]  Shi Zhou,et al.  The rich-club phenomenon in the Internet topology , 2003, IEEE Communications Letters.

[17]  Averill M. Law,et al.  Simulation Modeling and Analysis , 1982 .