Global computer networks are crucial to many businesses and the continued information explosion of the last decade ensures that this dependence will increase. Distributed processing systems if properly managed offer many attractive benefits beyond that of a standard network. However network architectures such as Open Systems Interconnection (OSI) and distributed system architectures such as Open Distributed Processing (ODP) are critically dependent on Quality of Service (QoS).
A pure definition for Quality of Service is almost certainly unattainable. It is difficult to separate the concepts of behaviour, performance, constraint and quality of service. Quality of Service is a term applied to a range of aspects of digital communications and distributed computing. These aspects include reliability, throughput, security, performance and usability, supported by functional aspects such as error detection and error correction. Quality of Service essentially sets the expectations and governs the performance of communications within a system. Much effort is now being aimed at ODP, especially multimedia applications where Quality of Service aspects such as end-to-end delivery and inter/intra network synchronisation are of increasing importance.
The SQUIDS project covers many wide ranging topics. Although not obvious in the title, SQUIDS includes implementation (or rather, support) of Quality of Service. A subsidiary project was undertaken on Quality of Service using User Analysis and a Testing Approach (QUANTA).
The Stirling Quality of Service Integrated Reference Model (SQUIRM) includes support components that the designer of a distributed application might reasonably expect in a distributed environment. Five support components are planned:
The coordinating service was prototyped in Java.
Up one level to Ken Turner - Research Projects
Last Update: 15th July 2006