Self-Adaptive System Case-Study of Architecture-Based Software Reliability
Authors
Abstract
In the last two decades different methods to assess reliability from an architecture-based system were proposed. Surveys and systematic reviews done in the last decade, summarize those methods and provide information about their applicability and suitability. However, those surveys and reviews also identify several shortcomings that today still exist in the architecture-based reliability research. The lack of studies applied to real world scenarios and the nonexistence of a common case-study used to compare different methods, are important and unsettle reliability shortcomings. As a result, they lead to a poor method validation and to several questions that still need to be answered.We try to address some of these shortcomings by applying our work to a real case-study and compare our predicted reliability values with the ones extracted from the real sys- tem. In addition, we apply our method to a self-adaptive system to understand the evolution of an architecture which its major quality requirement is reliability.