Service-Oriented Computing (SOC) enables the composition of loosely coupledservice agents provided with varying Quality of Service (QoS) levels,effectively forming a multiagent system (MAS). Selecting a (near-)optimal setof services for a composition in terms of QoS is crucial when many functionallyequivalent services are available. As the number of distributed services,especially in the cloud, is rising rapidly, the impact of the network on theQoS keeps increasing. Despite this and opposed to most MAS approaches, currentservice approaches depend on a centralized architecture which cannot adapt tothe network. Thus, we propose a scalable distributed architecture composed of aflexible number of distributed control nodes. Our architecture requires nochanges to existing services and adapts from a centralized to a completelydistributed realization by adding control nodes as needed. Also, we propose anextended QoS aggregation algorithm that allows to accurately estimate networkQoS. Finally, we evaluate the benefits and optimality of our architecture in adistributed environment.
展开▼