Assessing and Improving Interoperability of Distributed Systems

Achieving interoperability of distributed systems offers means for the development of new and innovative business solutions. Interoperability allows the combination of existing services provided on different systems, into new or extended services. Such an integration can also increase the reliability of the provided service. However, achieving and assessing interoperability is a technical challenge that requires high effort regarding time and costs. The reasons are manifold and include differing implementations of standards as well as the provision of proprietary interfaces. The implementations need to be engineered to be interoperable. Techniques that assess and improve interoperability systematically are required. For the assurance of reliable interoperation between systems, interoperability needs to be assessed and improved in a systematic manner. To this aim, we present the Interoperability Assessment and Improvement (IAI) process, which describes in three phases how interoperability of distributed homogeneous and heterogeneous systems can be improved and assessed systematically. The interoperability assessment is achieved by means of interoperability testing, which is typically performed manually. For the automation of interoperability test execution, we present a new methodology including a generic development process for a complete and automated interoperability test system. This methodology provides means for a formalized and systematic assessment of systems' interoperability in an automated manner. Compared to manual interoperability testing, the application of our methodology has the following benefits: wider test coverage, consistent test execution, and test repeatability. We evaluate the IAI process and the methodology for automated interoperability testing in three case studies. Within the first case study, we instantiate the IAI process and the methodology for Internet Protocol Multimedia Subsystem (IMS) networks, which were previously assessed for interoperability only in a manual manner. Within the second and third case study, we apply the IAI process to assess and improve the interoperability of grid and cloud computing systems. Their interoperability assessment and improvement is challenging, since cloud and grid systems are, in contrast to IMS networks, heterogeneous. We develop integration and interoperability solutions for grids and Infrastructure as a Service (IaaS) clouds as well as for grids and Platform as a Service (PaaS) clouds. These solutions are unique and foster complementary usage of grids and clouds, simplified migration of grid applications into the cloud, as well as efficient resource utilization. In addition, we assess the interoperability of the grid-cloud interoperability solutions. While the tests for grid-IaaS clouds are performed manually, we applied our methodology for automated interoperability testing for the assessment of interoperability to grid-PaaS cloud interoperability successfully. These interoperability assessments are unique in the grid-cloud community and provide a basis for the development of standardized interfaces improving the interoperability between grids and clouds.