Software quality is an important issue in software engineering. The evaluation of software architecture is crucial to achieve quality scenarios, which reduces development and maintenance costs. This work presents a discrete event simulation environment for the software architecture assessment considering both functional and quality aspects. Discrete event system specification (DEVS) formalism and the underlying framework are used to specify the simulation elements. DEVS is based on the system theory and follows the engineering and object-oriented fundamentals to construct complex dynamic systems. The proposed environment is built in a modular and hierarchical way that provides scalability and reusability advantages. Although the proposal is focused on three quality attributes and a few metrics, this approach enables the definition of new elements and metrics related to other quality attributes that can be visible at runtime. A traditional architecture illustrates the proposal, where preliminary computational experiments for this real software system and concrete quality scenarios demonstrate the feasibility of the integrated simulation environment for the software architecture evaluation. In addition, a discussion shows how the results could help architects make design decisions to improve software quality during development.
A discrete event simulation model for evaluating quality attributes, employing the software architecture, is proposed in this work. A metamodel of the software architecture domain that includes the concepts required for measuring quality attributes at runtime is specified. So, a simulation model is built from it, following the principles of hierarchy and modularity, assembling simple blocks to obtain complex blocks. DEVS framework is applied to obtain a decoupled model from the simulator, and the DEVS formalism is used to specify the elements of the simulation model. The objective of this approach is to provide information about the quality attributes that can be measured at runtime, introducing the discrete event simulation in the context of the software architecture design. This quantitative information will assist the architect to make decisions about the design of the system.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.