An Adaptive Control Strategy for Managing Concurrent Service Requests in Service-Based Systems

Service-Oriented Architecture (SOA) has attracted substantial research effort over the past decade. In service-based systems, where multiple services providing the same functionality are available, it is necessary to select the appropriate services to be used in service composition to provide better quality of service (QoS) for service consumers. In order to select the best service composition for concurrent service requests, we present an adaptive control strategy which performs service selection based on the historical usage information of all the candidate services. A case study of a data stream management system is presented to validate the effectiveness of the adaptive control strategy, and examine the effect of the concurrency level on the process of the strategy's decision.