The purpose of abstraction is not to be vague, but to create a new semantic level in which one can be absolutely precise....
- E. W. Dijkstra - from the Humble Programmer, 1972.
Workshop Programme: May 3, 2010
Chair Carlo Ghezzi
|Marta Kwiatkowska. "Challenges in quantitative verification for adaptive systems"|
Morning Session, Chair Lars Grunske
|-Leslie Cheung, Leana Golubchik and Nenad Medvidovic. "SHARP: A Scalable Approach to Architecture-Level Reliability Prediction of Concurrent Systems"
-Vittorio Cortellessa, Antinisca Di Marco, Romina Eramo, Alfonso Pierantonio and Catia Trubiani. "Digging into UML models to remove performance antipatterns"
-Ivo Krka, Leana Golubchik and Nenad Medvidovic. "Capturing Operational Profile with Probabilistic Automata for Architecture-Based Reliability Assessment"
Afternoon Session, Chair Marta Kwiatkowska
|-Husain Aljazzar, Matthias Kuntz, Florian Leitner-Fischer and Stefan Leue. "Directed and Heuristic Counterexample Generation for Probabilistic Model Checking - A Comparative Evaluation"
-Esteban Pavese, Victor Braberman and Sebastian Uchitel. "My Model Checker Died! How Well Did It Do?"
Session on ongoing projects, Chair Vittorio Cortellessa
|-Steffen Becker: "Trading off Quality Properties for Evolving Software Systems"
-Giordano Tamburrelli: "On the role of quantitative modelling and analysis in self-managing situational computing"
-Lars Grunske and Indika Meedeniya: "Archeopterix: A Tool Environment for Architecture Optimisation Based on Efficient Quantitative Evaluation"
Workshop Theme, Goals and Relevance
Modern software systems are increasingly complex and pervasive. They are often offered to clients as services, which must meet high-level expectations in terms of predictability. They are embedded in a fluid environment that is constantly evolving, because of changes in the requirements, in the operating environment, and in usage profiles. Traditional software development processes and methods are unable to cope with the challenges posed by this setting.
State-of-the-art model-driven approaches, in particular, emphasize the use of qualitative models in the development stage and systematic transformations that generate functionally correct implementations starting from well-defined requirements. If the need for evolution arises, the system undergoes suitable re-design, re-development, and re-deployment activities. These new settings call for quantitative as well as qualitative models and analysis techniques, techniques to collect and analyze data in the operational environment, and reasoning support to generate the necessary changes.
The needed quantitative models must support the reasoning about quality attributes such as performance, reliability, security. Different kinds of stochastic modeling techniques have been proposed to deal with these problems, and different approaches to analysis and verification are available. The use of models allows the prediction of the system quality before it is built and the understanding of the main effects of an architecture with respect to quality requirements. This prediction can be exploited to drive decisions about how to architect a software application so as to meet the quality requirements imposed on the design. Because of the limited a-priori knowledge about real-world behaviors and because of the likely changes in operational environments, however, models must evolve. Both their parameters and even their structure are likely to change, and the change in the models may imply further changes in the implementation. Thus, models must be kept alive at run time, and must be continuously refined to achieve increasingly better accuracy, by updating the relevant parameters and/or the model architecture. The workshop aim is to set up a multi-disciplinary community bringing together top researchers and practitioners belonging to different fields spanning:
- software design
- model-based development
- quality assessment
- quality prediction
- software evolution
- service-based development
The workshop co-location with ICSE 2010, which will be attended by numerous representatives from both software academia and industry, would provide a unique opportunity for researchers of different field to meet with practitioners.
Main Topics (non exclusive)
- formal definition of quality requirements
- languages for software design modelling including quality characteristic evaluation
- quality attribute models (such as performance, dependability, power consumption).
- integration of quantitative models into model-driven approaches
- quality testing, monitoring, measurement, and experimental design
- probabilistic verification
- statistical forecasting of quality attributes
- quantitative models at runtime
- quality requirements and software design evolution
- empirical validation of testing, prototyping models, simulation for assessing design quality
- design decisions and their quality impacts
- quality and software design governance
|31-01-2010 (Extended)||Paper Submission|
Submitted papers will be reviewed by the international Program Committee and accepted on their scientific merit and relevance to the topics of the workshop. Accepted papers will be published in the ICSE companion volume. Papers should not exceed 8 pages ACM proceedings format (10pt, single-space, double-column) and include an abstract of up to 150 words. Papers must not have been previously published or submitted elsewhere. If accepted, the paper must be personally presented at the workshop by one of the co-authors. Paper submission system is available online via Easychair