Performance Analysis of Weighted Fair Queues with Variable Service Rates

This paper provides an analytical technique to estimate the queue length and delay distributions for weighted fair queues (WFQ) with more than two queues and time-correlated variable service rates, based on two-dimensional decoupling. At first, temporal decomposition is used to convert the time-correlated queuing problem into a set of sub-problems over several time scales. Subsequently, queue decomposition exploits the queue weight dependencies to convert a multi-queue problem into a set of single-queue problems. The core of the analysis lies in estimating the multi-scale service rate models for each of these queues. The paper shows the hierarchy of this estimation and the dependency of the queue service rate on the other queues unused capacity and their weights. Simulation and analytical results on queue and delay survivor functions are in a good agreement