Network Protocols and the Causes of Chaotic Traffic

最近的实证研究表明，基于混沌理论的自相似流量模型可能比通常使用的网络流量模型更合适。如果不了解这种自相似行为的原因，就很难说这些模型的应用范围有多广，或者如果应该做些什么来减轻网络协议本身是否会使流量在工程兴趣的时间尺度上出现混乱，如果是这样，协议修改是否会有所帮助。这将通过模拟各种网络，现在和未来的现实模型来实现。在每个模拟中，将采用一个行为良好（非混沌）的输入流量模型，并将确定网络协议是否可以将这种行为良好的输入转换为混沌或自相似的流量。

Recent empirical studies have shown that self-similar traffic models, which are rooted in chaos theory, may be more appropriate than the models typically in use for network traffic. Without an understanding of the causes of this self-similar behavior, it is difficult to say how widely applicable these models are, or what if anything should be done to mitigate the will explore whether network protocols themselves can make traffic appear chaotic over time scales of engineering interest, and if so, whether protocol modifications might help. This will be achieved by simulating realistic models of a variety of networks, present and future. In each simulation, a well-behaved (non-chaotic) model for input traffic will be employed, and it will be determined whether the network protocols can turn this well-behaved input into chaotic or self-similar traffic flows.