Institutional Communication Service
Relying on high quality software systems is a priority of everyday life. Prof. Pezzè’s research group has been recently awarded with two Distinguished Paper Awards (assigned to the most relevant papers presented at a scientific conference) on the reliability of complex systems.
The first distinguished paper award has been granted at the International Conference on Software Engineering (ICSE), the premier international software engineering conference. The article (written by Cristina Monni and Mauro Pezzè) was awarded in in the NIER (New Ideas and Emerging Results) track, and presents the innovative idea to apply models and methods that recently emerged in statistical mechanics to tackle software engineering problems. The original intuition comes from the observation of the analogy between software failures and physical phenomena. In both cases we are interested in studying special states, failures in software systems, superconductivity in physical systems. The work is the most recent result of a four-year project carried out thanks to corporate funding, which has also led to the design of a product that has been recently presented on the market.
The second distinguished paper award has been granted at the International Conference on Software Testing (ICST) (another important conference in the field) to a paper written by Valerio Terragni and Mauro Pezzè. The paper presents a new technique for testing concurrent software systems. Concurrency, that is, the simultaneous execution of instructions in different streams, is an almost universal feature in modern software systems. It takes full advantage of modern computer parallel architectures, but introduces new types of failures that may lead to serious consequences. Testing such systems is extremely complex, and the paper presents a new technique to automate the testing activity, thus improving the overall quality and reliability of the system. The research is part of a project funded by the Swiss National Science Foundation for over one million francs. The project aims to build an automated platform for testing modern complex systems, where software, devices (sensors, telephones, mechanical parts and more), automated learning components, and people interactively adapt the overall system behavior to match emerging and relevant requirements.
The two projects are part of the research carried out by Prof. Pezzè’s group, born with the Faculty of Informatics in 2006. The group has hosted more than 20 PhD students and raised over 4.5 million francs in competitive funding. For more information: http://star.inf.usi.ch/star/