An Adaptive Heuristic Approach to Service Selection Problems in Dynamic Distributed Systems. Beran, P. P., Vinek, E., Schikuta, E., & Leitner, M. In Proceedings of the IEEE/ACM International Workshop on Grid Computing, pages 66–75, 2012. IEEE Computer Society.
An Adaptive Heuristic Approach to Service Selection Problems in Dynamic Distributed Systems [link]Paper  doi  abstract   bibtex   
Quality-of-Service (QoS) aware service selectionproblems are a crucial issue in both Grids and distributed, service-oriented systems. When several implementations perservice exist, one has to be selected for each workflow step. Several heuristics have been proposed, including blackboardand genetic algorithms. Their applicability and performancehas already been assessed for static systems. In order to coverreal-world scenarios, the approaches are required to deal withdynamics of distributed systems. In this paper, we proposea representation of these dynamic aspects and enhance ouralgorithms to efficiently capture them. The algorithms areevaluated in terms of scalability and runtime performance, taking into account their adaptability to system changes. Bycombining both algorithms, we envision a global approach toQoS-aware service selection applicable to static and dynamicsystems. We prove our hypothesis by deploying the algorithmsin a Cloud environment (Google App Engine) that allows tosimulate and evaluate different system configurations.
@inproceedings{BeranVSL_adaptive_2012,
	title = {An Adaptive Heuristic Approach to Service Selection Problems in Dynamic Distributed Systems},
	issn = {1550-5510},
	doi = {10.1109/Grid.2012.26},
	booktitle = {Proceedings of the {IEEE/ACM} International Workshop on Grid Computing},
	author = {Beran, Peter Paul and Vinek, Elisabeth and Schikuta, Erich and Leitner, Maria},
	year = {2012},
	publisher = {{IEEE} Computer Society},
	url = {http://doi.ieeecomputersociety.org/10.1109/Grid.2012.26},
	abstract = {Quality-of-Service (QoS) aware service selectionproblems are a crucial issue in both Grids and distributed, service-oriented systems. When several implementations perservice exist, one has to be selected for each workflow step. Several heuristics have been proposed, including blackboardand genetic algorithms. Their applicability and performancehas already been assessed for static systems. In order to coverreal-world scenarios, the approaches are required to deal withdynamics of distributed systems. In this paper, we proposea representation of these dynamic aspects and enhance ouralgorithms to efficiently capture them. The algorithms areevaluated in terms of scalability and runtime performance, taking into account their adaptability to system changes. Bycombining both algorithms, we envision a global approach toQoS-aware service selection applicable to static and dynamicsystems. We prove our hypothesis by deploying the algorithmsin a Cloud environment (Google App Engine) that allows tosimulate and evaluate different system configurations.},
	keywords = {distributed systems},
	pages = {66--75}
}

Downloads: 0