Edinburgh Napier University

  Cloud computing is a disruptive technology that aims to provide a utility approach to computing, where users can obtain their required computing resources without investment in infrastructure, computing platforms or services. Cloud computing resources can be obtained from a number internal or external sources. The heterogeneity of cloud service provision makes comparison of services difficult, with further complexity being introduced by a number of provision approaches such as reserved purchase, on-demand provisioning and spot markets.
The aim of the research was to develop a semantic framework for cloud computing services which incorporated Cloud Service Agreements, requirements, pricing and Benefits Management.
The proposed approach sees the development of an integrated framework where Cloud Service Agreements describe the relationship between cloud service providers and cloud service users. Requirements are developed from agreements and can use the concepts, relationships and assertions provided as requirements. Pricing in turn is established from requirements. Benefits Management is pervasive across the semantic framework developed.
The methods used were to provide a comprehensive review of literature to establish a good theoretical basis for the research undertaken. Then problem solving ontology was developed that defined concepts and relationships for the proposed semantic framework. A number of case studies were used to populate the developed ontology with assertions. Reasoning was used to test the framework was correct.
The results produced were a proposed framework of concepts, relationships and assertions for a cloud service descriptions, which are presented as ontology in textual and graphical form. Several parts of the ontology were published on public ontology platforms and, in journal and conference papers.
The original contribution to knowledge is seen in the results produced. The proposed framework provides the foundations for development of a unified semantic framework for cloud computing service description and has been used by other researchers developing semantic cloud service description.
In the area of Cloud Service Agreements a full coverage of the documents described by major standards organisations have been encoded into the framework. Requirements have been modelled as a unique multilevel semantic representation. Pricing of cloud services has been developed using semantic description that can be mapped to requirements. The existing Benefits Management approach has been reimplemented using semantic description.
In conclusion a proposed framework has been developed that allows the semantic description of cloud computing services. This approach provides greater expression than simplistic frameworks that use mathematical formulas or models with simple relationships between concepts. The proposed framework is limited to a narrow area of service description and requires expansion to be viable in a commercial setting.
Further work sees the development of software toolsets based on the semantic description developed to realise a viable product for mapping high level cloud service requirements to low level cloud resources.