Trade-off Optimisation of Web Service Composition Using Nature-Inspired Meta-heuristics
Dynamic service composition promotes the on-the-fly creation of value-added applications by combining existing service components. Large scale, highly dynamic service environments, like applications in the pervasive computing domain, pose many obstacles to service composition. Managing applications composed by loosely coupled autonomous components exhibits many interesting challenges. This research project attempts to address the question: How can we provide optimal and flexible composition configurations during run-time in the context of highly dynamic distributed environments?
We aim at formulating the above problem as a multi-objective optimisation problem and provide an efficient and quick optimisation engine. More specifically, given a set of services and a set of service composition plans, the goal of the desired optimiser is to find the composition configurations that optimise the overall performance based on dynamic objectives and preferences of the parties involved. To achieve this goal, the composition system should take into account not only the quality of each service component, but also how services are coordinated, communicate and interact within the composition.
Unexpected changes in various parameters of the environment may occur during the execution of a service composition and may affect not only the corresponding service component in isolation, but the end-to-end quality of the overall composed application. Consequently, the composition system should flexibly react and adapt to dynamic changes towards maintaining optimality during run-time. The objective of our research is to provide an algorithmic solution for supporting optimal, adaptive service composition in environments where dynamism is the key concern.
Main research interests:
- Cloud Computing
- Evolutionary Computation
- Ad-hoc and Sensor networks
- Probabilistic and Distributed algorithms
- Design and Analysis of Algorithms
- Developing Applications on the Internet
Efstathiou D., McBurney P., Zschaler S. and Bourcier J.: Surrogate-Assisted Optimisation of Composite Applications in Mobile Ad hoc Networks. In Proc. of the 16th Annual Genetic and Evolutionary Computation Conference (GECCO 2014), Vancouver, Canada, July 12-16, 2014, PDF
Efstathiou D., McBurney P., Zschaler S. and Bourcier J.: Efficient Multi-Objective Optimisation of Service Compositions in Mobile Ad hoc Networks Using Lightweight Surrogate Models. On the Special Issue on Adaptive Services for the Future Internet of the Journal of Universal Computer Science (JUCS), 2014 (to appear).
Efstathiou D., Williams J.R. and Zschaler S.: Crepe Complete: Multi-objective optimisation for your models. In Proc. 1st Int’l Workshop on Combining Modelling with Search- and Example-Based Approaches (CMSEBA’14), (co-located with MODELS), September 28, Valencia, Spain, 2014 (to appear).
Efstathiou D., McBurney P., Zschaler S., Bourcier J.: Flexible QoS-Aware Service Composition in Highly Heterogeneous and Dynamic Service-Based Systems. In Proc. of the 9th IEEE Int’l Conf. on Wireless and Mobile Computing, Networking and Communications (WiMob 2013), 7-9 Oct., Lyon, France, 2013, pp.592-599, LINK
Efstathiou D., McBurney P., Zschaler S. and Bourcier J.: Exploring Optimal Service Compositions in Highly Heterogeneous and Dynamic Service-Based Systems. In 5th Symposium on Search-Based Software Engineering, August 24-26, St. Petersburg, Russia, 2013, pp.312-317, LINK
Efstathiou D., McBurney P., Plouzeau N., Zschaler S.: Improving the Quality of Distributed Composite Service Applications. In Imperial College Computing Student Workshop September 27—28, London, United Kingdom, 2012, pp. 49-55, PDF