Informatica > Q-ImPrESS

Abstract del Progetto

The Q-ImPrESS project aims at bringing service orientation to critical software systems, such as industrial production control, telecommunication and critical enterprise applications. All these domains share a need for guaranteed end-to-end quality of service, but also a need to evolve over their long lifetimes. The Q-ImPrESS project targets this challenge by providing a method to allow developers, users and maintainers to foresee the impact of design decisions and evolutionary changes to the system not only on its overall quality of service, but also on its internal quality properties such as maintainability.
Therefore, a new service architecture model is developed which is accompanied by reengineering tools capable of extracting this model from existing source code. The service architecture model is a uniform intermediate representation used to simulate different evolutionary changes for the purpose of assessing their impact on the quality attributes of the system. Using the methodology for trade-off analysis which will be developed in Q-ImPrESS, the software engineers will be able to take informed design decisions before even writing a single piece of code and thus save costly development effort and time. The Q-ImPrESS methodology will ensure that the resulting software systems are more predictable and it will alleviate the risk of design changes affecting quality attributes.

Responsabile del progetto

Sito del progetto

Risultati del progetto ed eventuali pubblicazioni scientifiche/brevetti

The Q-ImPrESS project aims at defining:
  • A novel service model that augments the service description with the information required to make predictions on the QoS of services and service compositions. Under QoS, we explicitly understand performance and reliability metrics. Once this model is established and thoroughly validated, it is planned to submit parts of it to standardization bodies to extend current existing standards.
  • An impact analysis technique that advances current QoS prediction methods by making them more service specific.

The advancements to current quality impact analysis techniques will be evaluated through experiments and measurements, whose results will be documented in project deliverables.