The focus of this research is on problems that arise in the context of QoS enabled IP networks and their use by applications. In particular, the main theme of the research, is the development of a greater synergy between potential users of network QoS capabilities, i.e., applications, and the design and implementations of those QoS capabilities (signaling, scheduling and access algorithms) within the network. In order to accomplish such a goal we are looking specifically at the use and selection of network QoS services and parameters by applications. The goal here is to identify performance parameters that are of most significance to applications, and translate that knowledge into utility curves that can then be used by both applications to select the appropriate network service, and by the network to make intelligent decisions on how to best handle an application's traffic in case of resource contention. This includes not only assessing sensitivity to traditional QoS parameters such as bandwidth, loss, delay, and jitter of different types of applications traffic, e.g., transactions, audio, video, etc., but also considering more complex scenarios involving applications with multiple traffic streams which might involve different (dynamic) resource sharing rules depending on the availability of network resources.
The research focuses on the identification of both network and end system QoS capabilities, that are explicitly aimed at better application support. This requires an iterative process between applications and the network in order to identify applications requirements and determine how they can be best met by the network. Of particular interest in the context of this research, is to develop an understanding of which engineering support is most useful to applications when accessing a QoS enabled IP. The investigation of these issues is being carried out using a rich multimedia client-server application that combines multiple types of streams, i.e., audio, video, and data. The benefit of using such an application is not only that the different requirements of its individual components exercise a wide range of network QoS capabilities, but also that the dependencies that exist between streams create a new set of resource sharing requirements. For example, the choice of which streams to degrade and how degradation should be applied across streams in the presence of congestion, is likely to depend not only on the level of congestion, but also on application level semantic.
Another very important aspect of the research relates to the pricing of services and the impact that it will have on user behaviour. For that one needs to understand the range of possibilities presented by the multimedia services in conjunction with the capabilities of the end system, and how that can be presented to the end user who will make a choice based upon availability, need and price.
In order to gain an understanding of these issues, and develop and test the necessary network mechanisms, this research relies heavily on experiments. Experimentation will take place in the context of a lab devoted to multimedia and networking, that includes end-systems as sources of application traffic, and networking equipment made available by several vendors. Furthermore, work on the design and implementation of new mechanisms to better support application service requirements, will and must be done in close collaboration with equipment vendors so as to facilitate their incorporation and testing on the available platforms.