The Problem
Modern, integrated computer systems are typically built using a service-oriented architecture (SOA). They consist of a collection of software services to ensure the on-demand delivery of information within the organisation or to customers. Because demands on systems can be unpredictable, the performance of some services can degrade or even fail outright. At best this frustrates everyone, from customers to management and systems administrators. At worst it can be disastrous, causing political or brand embarrassment and/or a mass exodus of customers.
A range of technologies already exist to provide system monitoring and alert you when things fail. This is fine if you have the time to wait for your systems engineers to trace the problem and re-boot the right servers. But organisations need to provide continuous service surety in the face of changing demand in real-time.
The Solution
The Adaptation Engine technology provides a plug-in which is an autonomic nervous system for SOA environments. It can help organisations sense performance problems before a disaster arises, and adapt in real-time to the ongoing changes in load. End-users of the services need not notice anything except consistent service levels.
The Adaptation Engine technology is applied non-intrusively and works at the boundaries of integrated services, so no radical alterations to existing systems or their integration points are required.
Publications
The best overview of the technology is given in this paper:
- Liu, Y., Gorton, I., Trivedi, N. A Lightweight and Extensible Architecture for Adaptive Server Applications, Software: Practice and Experience, 38(8), Wiley, 2007, pp. 853-883.
Some more recent work is available here:
- Liu, Y., Tan, M., Gorton, I., Clayphan, J. A. An Autonomic Middleware Solution for Coordinating Multiple QoS Controls. Proceedings of the 6th International Conference on Service Oriented Computing, Springer, LNCS 5364/2008, 2008, pp. 225-240.
Work on this technology follows on from the previous AMP project - the earlier publications are available here.
Download
- One-page summary flyer (PDF).