MPLS DiffServ-Enabled Traffic Engineering: A Scalable QoS Model for Optical-Speed Media Streaming Networks

Abstract

The rapid growth of the Internet and the widespread deployment of networks built around the Internet Protocol suite are creating a demand for new capabilities in IP networks. In order to support the multitude of emerging multimedia applications such as streaming video, voice over IP, distance learning and other real-time services, it is necessary for the Internet to support, in addition to best-effort service, proper service-level agreements that guarantee a specified level of throughput, timing and network reliability, irrespective of the usage level and individual network failures. In detail, multimedia applications with real-time properties require tight guarantees in terms of delay, delay variation (jitter), bandwidth, packet loss, and availability that usually identify some service classes and specify the forwarding treatment, in term of QoS due to traffic belonging to each service class. To meet these requirements, the network must be enhanced with new technologies that offer adequate capabilities for controlling its behavior. Together, the capabilities offered by the combination of Differentiated Services and Multi-Protocol Label Switching (MPLS) enhance the ability of the network operator to control the network to deliver service according to the performance, QoS and scalability needs of the near future delay-sensitive multimedia networking applications. This paper presents and evaluates a complete framework of MPLS DiffServ facilities for deployment of performance-aware multimedia streaming networks.