Decision Model for Software Architectural Tactics Selection Based on Quality Attributes Requirements

Abstract Due to increasing industrial demands toward software systems with increasing complexity and challenging quality requirements, software architecture and implementation mechanisms become an important activity. The decisions made during architecture design have significant implications on quality goals. As addressed, there is a lack of available standard models, architectures or frameworks for enabling implementation of quality attributes specially for business intelligence environment and applications in order to rapidly and efficiently supports decision-making. In addition, a lack of researches related to Quality Attributes (QA) requirements, its implementation tactics, and interrelations or correlations between them. The increasing systems complexity mandates software architects to choose from a growing number of design options (decisions) when searching for an optimal architecture design in a specific domain with respect to a defined (set of) quality attributes and constraints. This results in a design space search that is over human capabilities and makes the architectural design task more complicated. In this paper, researcher aimed to reveal most of quality attributes implementation tactics affecting applications architectures, properties. Several quality attributes of software investigated using applied research methods with mixed quantitative (linear) and non-linear analysis techniques. It proposes an initiative for finding an easy and systematic way of addressing quality attributes requirements to a set of implementing architectural tactics. Finally, the findings analyzed and visualized in a way that can support decision stakeholders in addition to a new concept of “safe-tactics” introduced as reduced (pruned) set of tactics that are claimed to be better used in general refactoring cases. In addition, a software tool is developed throughout this research effort as result of gained knowledge and addressing the research findings.