RDF-based IT Configuration Management Database

These days, a lot of organizations are highly dependent on their IT infrastructure. Many organizational service provisioning processes are depending on reliable and high-quality IT services. Therefore the area of IT Service Management (ITSM) has become a very important disciple. One of the most established ITSM approaches is the IT Infrastructure Library (ITIL) Version 3, a set of best practices for managing IT infrastructures. ITIL V3 proposes the concept of maintaining a Configuration Management System (CMS), which aggregates the data of multiple Configuration Management Databases (CMDB). This concept supports the management of heterogeneous IT environments as well as guaranteeing the integrity of the maintained infrastructure data. The developed Semantic Configuration Management Database (SeConD) framework provides the common functionalities required by a CMDB for managing IT infrastructures. In contrast to already existing open source solutions, the developed framework is based on a standardized data model; in this case the Common Information Model (CIM). The goal of this thesis is mapping the object-oriented CIM model to a semantic data model. This has been leading to the development of a CMDB framework which leverages Semantic Web technologies. Based on the resulting CIM ontology, a semantic CMDB framework has been established. The primary focus of the resulting CMDB lies in the area of storing IT-related Configuration Items (CI) provided by Web Based Enterprise Management (WBEM) enabled hosts. Therefore a Triplestore database provides the required storage back-end capabilities. The SeConD framework only implements the required functionalities in the scope of a single CMDB. Further CMS and business logic parts are not incorporated directly since they are out of the scope of this thesis. The SeConD framework is an enabler for further work in the area of semantic CMDB research. This allows examining the possibilities of using predicate logic for consistency purposes as well as business rule integration. Therefore further work would include the practical application of this framework in a smaller testbed to collect performance indicators for the reasoning process as well as the performance impact of increasing rule complexity.