Adaptive Agile Performance Modeling and Testing
Marek G. Stochel, Mariusz R. Wawrowski, James J. Waskiel · 2012
Rising industry trends in acceptance for Agile approaches in software development provides opportunities as well as challenges in the ever-changing environment of software development. On the one hand, fast feedback of working code combined with close customer collaboration enables increased software development productivity by providing greater accuracy of where to target the development effort. On the other hand, the growing interdependence of the product software components within complex systems gradually evolves considerable challenge in terms of assessing system performance and planning for future system expansion. In addition, and equally important, establishing and maintaining performance modeling becomes more and more costly due to the increasing volatility of requirements. We should also consider the growing complexity of systems being built, integration with third party subsystems, and at the same time the fact that engineering teams face tight project budgets as businesses are looking to reduce capital expenditures. To address these concerns, an adaptive performance modeling approach supported by automated performance analysis is proposed. This approach combines predictive engineering techniques, experimental feedback obtained through testing, and continuous data acquisition including knowledge-based assessments provided by the engineers.