TCP/IP traffic dynamics and network performance: A lesson in workload modeling, flow control, and trace-driven simulations

The main objective of this paper is to demonstrate in the context of a simp le TCP/IP-based network that depending on the underlying assumptions about the inherent nature of the dynamics of network traffic, very different conc lusions can be derived for a number of well-studied and apparently well-und erstood problems in the area of performance evaluation. For example, a traf fic workload model can either completely ignore the empirically observed hi gh variability at the TCP connection level (i.e., assume "infinite sources" ) or explicitly account for it with the help of heavy-tailed distributions for TCP connection sizes or durations, Based on detailed ns-2 simulation re sults, we illustrate that these two commonly-used traffic workload scenario s can give rise to fundamentally different buffer dynamics in IP routers. U sing a second set of ns-2 simulation experiments, we also illustrate a qual itatively very different queueing behavior within IP routers depending on w hether the traffic arriving at the router is assumed to be endogenous in na ture (i.e., a result of the "closed loop" nature of the feedback-based cong estion control algorithm of TCP) or exogenously determined (i.e., given by some conventional traffic model - a fixed "open loop" description of the tr affic as seen by the router).