A Proportional Jitter Scheduling and Buffer Management Algorithm for Differentiated Services Networks

This paper proposes a method for scheduling and buffer management in differentiated services networks which is based on JoBS algorithm. JoBS differentiates traffic classes base on delay and loss rate. Furthermore, it supports both relative and absolute constraints. Nowadays with the expansion of Internet traffic and its diversified service requests, the focus is on requirement of new traffic used on the Internet. The proposed method focuses on packet scheduling based on the jitter (variance of delay), since new traffics that will be mostly used in the next generation Internet are more susceptible to variance of delay. Drawing on concepts of jitter and the difference between scheduling base on jitter and delay, this paper analyzes the effectiveness of the proposed method by simulation examples that are presented here. Results show that proposed method can proportionally differentiated classes based on delay, loss rate and jitter. Also, it fulfills the absolute conditions which are defined by absolute parameters. This paper concludes with the suggestion that this method comparing to the JoBS can proportionally differentiate classes based on more parameters and its result of differentiation is better.