Multiple time scale TCP: Harnessing proactivity in self-similar burstiness
Teletraffic Engineering in the Internet EraProceedings of the International Teletraffic Congress - ITC-I7, том 4 из Teletraffic Science and Engineering, Elsevier, (2001)DOI: http://dx.doi.org/10.1016/S1388-3437(01)80140-2
Аннотация
Measurements of network traffic have shown that self-similarity is an ubiquitous phenomenon present across diverse network environments. In previous work, we studied the feasibility of exploiting long-range correlation structure in self-similar traffic in the context of rate-based congestion control. In this work, we extend the multiple time scale control framework to window-based congestion control, in particular, TCP. This is performed by interfacing \TCP\ with a large time scale control module which adjusts the aggressiveness of bandwidth consumption behavior exhibited by \TCP\ as a function of large time scale network state, i.e., information that exceeds the time horizon of the feedback loop as determined by RTT. How to effectively utilize such informationdue to its probabilistic nature, dispersion over multiple time scales, and affection on top of existing window-based congestion controlsis a nontrivial problem. We define a modular extension of TCPa function call with a simple interfacethat applies to various flavours of \TCP\ (e.g., Tahoe, Reno, Vegas) and show that it significantly improves throughput performance. We show that multiple time scale \TCP\ endows the underlying feedback control with proactivity by bridging the uncertainty gap associated with reactive controls which is exacerbated by the high delay-bandwidth product in broadband WANs. By harnessing large time scale correlation structure in network traffic, we are able to mitigate the reactive cost of TCP's feedback congestion control.