Nowadays, a typical software development process involves many developers which participate in the development process by using a wide variety of development tools. As a consequence, the data representing the project as a whole is distributed over different development tools. For the purpose of consistency, maintainability, and traceability it is an essential task to be aware of the relationships between semantic equivalent data in different tool repositories. The real-time systems Lab at the Technische Universitat Darmstadt performs research in the area of tool and metamodel integration to provide solutions to overcome this gap. In this demonstration we present the metamodeling framework MOFLON that addresses these issues by bringing together the latest OMG standards with graph transformations and triple graph grammars. Using MOFLON, developers can generate code for specific tools needed to perform analysis and transformation on one development tool or to incrementally integrate data of different modeling tools.

As an increasing number of industrial software development projects adopt model-driven development technologies, there is an emerging need for analysis, transformation, and integration of models. In this paper, we propose MOSL, an integrated specification language composed of MOF 2.0, OCL 2.0, Fujaba SDM, and Triple Graph Grammars as a suitable metamodeling and specification language. This language is the basis of the MOFLON metamodeling framework, which is currently developed at our department.

This paper describes an architecture analysis tool-set supporting evolutionary improvement of the software architecture of an existing medical imaging system. The toolset has been developed with co-operation between Philips Research and Philips Medical Systems. The medical imaging system is a large and software intensive system. Improvements to its architecture will lead to many development and business benefits, provided the improvements can be made efficiently and without major disruptions. The major challenge in supporting these improvements was finding a technical solution that fits well with the organization and processes of the development department. We came up with an approach that enables the migration to be performed in small increments and in line with existing processes and procedures established to meet requirements for patient safety and other qualities. Our tools provide relevant information about the current status and recent trends in software architecture, allowing software architects and engineers to monitor the progress of the migration and to identify potential problems as early as possible.

We show how two web-based architecture trend analysis tools, considering different views and their changes over time, contribute to our daily effort to cope with the complexity of a large, software-intensive medical imaging system.