QoS adaptation in real time systems based on CBWFQ

摘要

The evolution of the World Wide Web (WWW) service has incorporated new distributed multimedia conference applications, powering a new generation of e-services development, and allowing improved interactivity [1]. Most common interactive real-time applications are fault-tolerant but suffer from QoS limitations; low-latency requirements and reliability are cumulative to conference success [2], it may easily absorb all network resources and the subjective quality sensed by users would remain poor if the available resources are used indiscriminately. This paper simulates a network infrastructure design that classifies and prioritizes the real-time traffic in order to improve the performance of the real-time applications, taking into consideration of the performance of non real-time applications. This infrastructure has been designed to present a typical network of a university campus, so that in this way it can be implemented in any campus. In this paper 3 scenarios are suggested first before enabling any QoS mechanism at the routers, second use class based weighted fair queuing using low latency queue (CBWFQ-LLQ) for video traffic and last scenario apply CBWFQ-LLQ for voice. The simulation is conducted using OPNET IT Guru. Simulation results demonstrated that applying LLQ for voice improves the performance of overall real-time and non real-time application.