Using Blackboards to Optimize Grid Workflows with Respect to Quality Constraints
GCCW '06 Proceedings of the Fifth International Conference on Grid and Cooperative Computing Workshops
ACM Transactions on Autonomous and Adaptive Systems (TAAS)
Grid workflow optimization regarding dynamically changing resources and conditions
Concurrency and Computation: Practice & Experience - 2nd International Workshop on Workflow Management and Applications in Grid Environments (WaGe2007)
Future Generation Computer Systems
VieSLAF Framework: Enabling Adaptive and Versatile SLA-Management
GECON '09 Proceedings of the 6th International Workshop on Grid Economics and Business Models
A Parallel Branch and Bound Algorithm for Workflow QoS Optimization
ICPP '09 Proceedings of the 2009 International Conference on Parallel Processing
Web Services: Concepts, Architectures and Applications
Web Services: Concepts, Architectures and Applications
Service selection algorithms for composing complex services with multiple qos constraints
ICSOC'05 Proceedings of the Third international conference on Service-Oriented Computing
Bringing semantics to web services: the OWL-S approach
SWSWPC'04 Proceedings of the First international conference on Semantic Web Services and Web Process Composition
A Business Rules Driven Framework for Consumer-Provider Contracting of Web Services
Proceedings of International Conference on Information Integration and Web-based Applications & Services
Review: Cloud computing service composition: A systematic literature review
Expert Systems with Applications: An International Journal
Hi-index | 0.00 |
With the advent of Cloud computing, there is a high potential for third-party solution providers such as composite service providers, aggregators or resellers to tie together services from different clouds to fulfill the pay-per-use demands of their customers. Customer satisfaction which is primarily based on the fulfillment of user-centric objectives is a crucial success factor to excel in such a service market. The clients' requirements, if they change over time even after the desired solution composition, may result in a failure of this approach. On the other hand, business prospects expand with the possibility of reselling already designed solutions to different customers after the underlying services become available again. The service composition strategies must cope with the above-mentioned dynamic situations. In this paper we address these challenges in context with the customer-driven service selection. We present a formal approach to map customer requirements onto functional and non-functional attributes of the services. We define a happiness measure to guarantee user satisfaction and devise a parallelizable service composition algorithm to maximize this happiness measure. We devise a heuristic approach based on historical information of service composition to rapidly react to changes in client requirements at design time and indicate run-time remedies such as for service failures. The heuristic algorithm is also useful to recompose similar solutions for different clients with matching requirements. Our algorithms are evaluated by the results of a simulation developed on the workflow tool Kepler coupled with a C++ implementation of the optimization algorithms.