Use this URL to cite or link to this record in EThOS: http://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.566288
Title: Matching of service feature diagrams based on linear logic
Author: Naeem, Muhammad
Awarding Body: University of Leicester
Current Institution: University of Leicester
Date of Award: 2012
Availability of Full Text:
Access from EThOS:
Access from Institution:
Abstract:
Managing variability is essential for an efficient implementation of end-user services that can be customised to individual needs. Apart from variations in selection and orchestration, also third-party services may have to be customisable. Although feature diagrams provide a high-level visual notation for variability, their use for specifying variability of services raises the problem of matching a required feature diagram against a set of provided ones. In particular, the established interpretation of feature diagrams in Propositional Logic is not expressive enough for matching in the context of service variability. The problem becomes more visible when a certain requirement is going to be satisfied by a combination of multiple offers with overlapping features, which is a consequence of idempotence in Propositional Logic. To address this problem, we propose service feature diagrams with semantics in Linear Logic. Linear Logic only allows the use of idempotence on the propositions with modalities. The permissible selection of features of a service feature diagram is called an instance diagram. We provide rules to obtain instance diagrams from the service feature diagram. The semantics of instance diagrams are also supported by Linear Logic. This thesis not only introduces service feature diagrams, but also formalises their matching as linear deduction. We propose two categories of rules to verify diagrammatically if a collection of service descriptions satisfy the requirements. First, graphical matching rules are used to match service feature diagrams of requestor and provider. Second, graphical merging rules are used to merge multiple feature diagrams contributing to satisfy the requestor’s demands. We prove the correctness of these rules using the inference system of Linear Logic. We also provide the analysis of graphical rules and show that the application of the graphical rules is independent of the context in service feature diagram, i.e., graphical rule can be applied anywhere in a service feature diagram.
Supervisor: Heckel, Reiko; Fiadeiro, Jose Sponsor: Not available
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral
EThOS ID: uk.bl.ethos.566288  DOI: Not available
Share: