A Service Framework for Capability-based Adaptation in Adaptable Service Systems

Abstract

Networked services are considered. A networked service is offered by a service system consisting of components, which by their interworkings provide a service in the role of a service provider to users. An adaptable service system is a service system that is able to adapt to changes brought about by users, nodes, capabilities, system performance, and service functionalities. A capability is defined as an inherent property of a node, which is used as a basis to implement a service. This thesis focuses on adaptability aspects related to nodes and capabilities.

A service framework is a system for the specification, management, and execution of adaptable service systems. Functionalities of the service framework enable various adaptability types, as follows: capability-related adaptation, functionality-related adaptation, and context-related adaptation. The capability-related adaptation is considered in this thesis. Capability-related adaptation is related to the shortage of capabilities with appropriate logical functionality or the overload or failure of capabilities leading to reduced system performance.

This thesis presents a service framework for capability-related adaptation. This service framework consists of a computing architecture, concepts, ontologies, service component models, mechanisms, system architectures, and service execution platforms. The service framework presented is related to TAPAS (Telematics Architecture for Play-based Adaptable Service Systems). The TAPAS concepts and architectures that existed at the beginning of this Ph.D. study are the foundation of this thesis. On the other hand, the research work presented in this thesis has contributed to new TAPAS concepts, ontologies, service component models, mechanisms, system architectures, and service execution platforms.

The service framework is classified into six contibutions (C1-C6) as follows: C1: Ontology-based computing architectureC2: Ontologies, models, and representationsC3: Policy-based reasoningC4: Autonomic-based modelC5: Capability-related adaptation functionality realisationC6: Service execution platforms, prototypes, and simulations Ontology-based computing architecture is an architectural framework for the specification and execution of adaptable service systems that provide any service functionality. Ontologies, models, and representations are applied for the formalisation and representation of the necessary concepts; these are also the basis for the formal definition of the requirements of the capabilities and service component specifications, i.e., the policy and goal specifications. Policy-based reasoning is a support functionality that takes adaptation actions based on flexible and expressive behavioural specifications. The Autonomic-based model is composed of the specification and learning mechanism of the Autonomic Elements that make up an autonomic service system, which is a specialised version of adaptable service systems. Capability-related adaptation functionality realisation defines specific functionalities, interworkings, and structural organisation of the service components that implement the functionalities required for capability-related adaptation. Finally, service execution platforms, prototypes, and simulations provide the validation and evaluation of the remaining contributions