Evolution of the internet QoS and support for soft real-time applications

Abstract

I. INTERNET QUALITY OF SERVICEReal-time networked applications depend on network parameters such as bandwidth, delay, jitter (interpacket delay variation), and loss for their correct operation. The degree of tolerance or sensitivity to each of these parameters varies widely from one application to another. Critical applications, such as remote surgery and online trading systems, require reliability of such parameters as well as guaranteed delivery. Emerging applications such as home networking, intelligent appliances, factory supply-chain networks, as well as the vast majority of multimedia applications also require different levels of service quality from the underlying network in terms of these parameters. In traditional circuit-switched networks such as the telephone system, the quality of a connection can be measured and guaranteed in terms of connection setup delays, media quality (e.g., voice or video quality), and trunk availability. On the other hand, packet-switched networks, such as the current Internet, were not designed to provide per-connection service guarantees. As a result, packets of a particular flow can reach their destination via different paths and experience different amounts of delay, jitter, and loss along the way. This is a critical problem on the current Internet as more and more applications, originally developed for circuit-switched networks, are being migrated onto the Internet.The current Internet architecture is based on the "best-effort" service model, which has worked well for traditional applications such as e-mail, file transfer protocol (FTP), telnet, and hypertext transfer protocol (HTTP). Most of these applications are based on the transmission control protocol (TCP) suite, which provides reliable data delivery service without guaranteeing any delay and jitter bounds. On the other hand, supporting real-time and business-critical applications over a wide-area network requires classification of the application traffic and service-level guarantees from the underlying network for each class of traffic based on a number of factors. A number of proposals have been put forward to enhance the reliability and efficiency of the current Internet. A major objective of these proposals is to provide multiple service classes for either individual flows or flow aggregates with a quality that is suitable for these flows. Once implemented, such service classes will enable

A RED modified weighted moving average for soft real-time application

“…Real-time networked applications depend on network parameters such as bandwidth, delay, jitter (interpacket delay variation) and loss for their correct operation. The degree of tolerance or sensitivity to each of these parameters varies widely from one application to another (El-Gendy et al, 2003). However, real-time traffic is not so sensitive to packet losses, depending on the codecs and loss recovery schemes used at the receiver (Wang, 2006).…”

A Reliable and Robust QoS Routing Protocol for Mobile Wireless Sensor Networks

“…Quality of service (QoS) is defined as "the capability to provide resource assurance and service differentiation in a Network" [5]. Quality of Service (QoS) aims at providing better networking services over current technologies such as ATM, Ethernet and others.…”

Evaluation of delay performance of traffic shapers

“…Growth in the real-time applications on the Internet requires better performance from networks [5]. A slight drop in network performance may have a high impact on the tasks being performed due to high sensitivity of the application.…”