Jan Jelschen

Towards Applying Reengineering Services to Energy-Efficient Applications

Conserving resources and saving energy has become an important issue for information and communication technology. With increasing adoption of smart phones and tablet PCs, reducing energy consumption in mobile computing is of particular significance. User expectations towards their mobile devices are rising, and functionality is increasing. Accordingly, available energy is made a scarce resource. This paper discusses how software reengineering techniques, like dynamic analysis and refactoring, can be applied to the field of energy-aware computing, to monitor, analyze, and optimize the energy profile of mobile applications and devices.

SENSEI: Software evolution service integration

Software evolution tools mostly implement a single technique to assist in achieving a specific objective. Overhauling, renovating, or migrating large and complex legacy software systems require the proper combination of several different techniques appropriate for each subtask. Since few tools are built for interoperability, the setup of a toolchain supporting a given software evolution process is an elaborate, time-consuming, error-prone, and redundant endeavor, which yields brittle and inflexible toolchains with little to no reusability. This paper presents SENSEI, an approach to enable the implementation of an integration framework for software evolution tools using component-based, service-oriented, and model-driven methods, to ease toolchain creation and enable agile execution of software evolution projects. It will be evaluated by implementing and using it to build the toolchains supporting two software evolution projects, and having practitioners assess its usefulness.

Refactorings for Energy-Efficiency

Energy-efficiency is an important topic in information and communication technology and has to be considered for mobile devices, in particular. On the one hand, the environment should be protected by consuming less energy. On the other hand, users are also interested in more functionality of their mobile devices on hardware and software side, and at the same time, longer battery durations are expected. Hence, the energy consumption for mobile devices should be more efficient. This paper shows an approach to save energy on application level on mobile devices. This approach includes the definition, detection, and restructuring of energy-inefficient code parts within apps. Energy savings are validated by different software-based energy measurement techniques.

Describing Modeling Deltas by Model Transformation

Since large scaled software models typically exist in many revisions, extraction and representation of differences between versions is a crucial issue of model version systems. While handling model differences is playing an important role in evolution of models, there is a need for appropriate techniques to represent model differences. This paper shows a meta-model-generic and transformation based approach to the representation of model differences. Domain specific languages are generated to represent model differences. Differences are mapped to a set of model transformation rules. To demonstrate the approach, it is applied to activity diagrams.

A Toolchain for Metrics-based Comparison of COBOL and Migrated Java Systems

Migrating COBOL legacy systems to Java results in functional equivalent systems expressed in the new language, while the programming paradigm remains that of the old systems. The quality of translated code is therefore assumed to be inferior, if held to the standards of the target language’s paradigm. This paper presents an integrated toolchain enabling metricsbased comparisons of original and translated systems to substantiate or refute this hypothesis, characterize the change in code quality, and gain insights for the improvement of translation tools.