Efficient fair queueing using deficit round-robin
IEEE/ACM Transactions on Networking (TON)
Statistical properties of MPEG video traffic and their impact on traffic modeling in ATM systems
LCN '95 Proceedings of the 20th Annual IEEE Conference on Local Computer Networks
Simulation with Arena
Hi-index | 0.00 |
Traffic management in communication environments like the Internet is vital in meeting the demand of users. The recent introduction of various dynamic scheduling algorithms has been to target improvements in QoS (Quality of Service). These new algorithms, such as dual queueing, aim to give better QoS to most traffic at the expense of a few rather than fairly give poor QoS to all customers. This is achieved through scheduling decisions that can change depending on the traffic conditions. The MPDQ introduces different classes into this type of scheme with the aim of enhancing higher-class packets QoS without great expense to lower class packets. In this paper we analyse the differences in loss and waiting time between preemptive and non-preemptive service disciplines for the MPDQ and then identify the best queueing regime for the pre-emptive MPDQ. We show that non-preemptive service dispensing is superior to packets of two classes than the preemptive scheme, and that highest class first (HCF) is the best queueing regime for either of these algorithms.