Weighted Round Robin and Rate Limiter based Fair Queuing for WRR

Abstract

In this paper we present a new packet scheduling method based on parallel usage of multiple WRR schedulers, rate limiters and output bandwidth calculation for modern NGN networks. The main idea of the presented method is to provide queueing fairness within queues. The method provides the same results in output bandwidth allocation as the compared algorithm, while within one queue flows with different packet size and arrival rates gets the same output bandwidth. With this method we are able to achieve the overall result of bandwidth assignment as algorithms like WRR, WFQ, WRRPQ and LLQ by only changing the mathematical model used to calculate the bandwidth assignment. We call this method Weighted Round Robin and Rate Limiter based Fair Queuing (WRRRLbFQ). We prove the model outcome with simulation results using NS2 simulator and compare the behavior with the WRR scheduler.