Fluid Methods for Modeling Large, Heterogeneous Networks

Abstract

REPORT DOCUMENTATION PAGEForm Public reporting burden for this collection of information is estimated to average 1 hour per response, including the time for reviewing instructions, searching existing data sources, gathering and maintaining the data needed, and completing and reviewing this collection of information. Researchers from the University of Massachusetts developed fluid-based methodologies for characterizing the behavior of large IP networks handling large numbers of TCP and UDP flows. These methodologies provide for rapid and efficient rates of individual and aggregate flows. These fluid models were also applied to the problems of characterizing the spread of worms and viruses and to the cascade of failures within the BGP routing infrastructure. The resulting fluid models were used to develop novel active queue management mechanisms resulting in more stable TCP performance and novel rate controllers for the purpose of providing minimum rate guarantees to TCP flow aggregates. Last, methodologies and tools were developed to integrate fluid network simulation with packet-level simulation.

Bulletin of the Polish Academy of Sciences: Technical Sciences

“…The maximum values of q and W are dependent on the buffer capacity and maximum size of the window. Such a Fluid Flow model may be extended to any network topology with any number of TCP flows [52]. This method allows us to evaluate the behaviour of TCP/AQM mechanisms for large networks.…”

Diffusion Approximation Model of TCP NewReno Congestion Control Mechanism

“…The maximum values of q and W depend on the buffer capacity and maximum window size. This fluid flow model may be extended to any network topology with any number of TCP flows [39]. The article [32] presents a numerical example for a real network topology with 134 023 nodes and 50 000 flows.…”