Um processo para monitoração de requisitos de qualidade de software utilizando informações arquiteturais

Abstract

Advances in building computer systems allowed the development of increasingly sophisticated software. In this context, software developers are faced with complex problems that can be difficult to resolve. One of the fundamental steps to build a software is to define its main features, that is, define the requirements that can be satisfied, which is not an easy task. Typically, requirements are classified as functional and non-functional. The non-functional requirements define system’s properties or how it should operate. They are also called quality requirements, since they are important to ensure that the system is feasible to use and meets the needs of users. Despite their importance, they are difficult to assess and be measured in order to generate evidence of its full service. One reason for this difficulty is the need to evaluate the software at run-time, which is a complex task that involves large volumes of data to be analyzed. This work aims to systematize and facilitate the monitoring of quality requirements from characteristics of architectural styles used in the software architecture design. In this sense, the research question that guided this study was “how to monitor and analyse quality requirements for a software system”. On that basis, it was proposed a process, called ArMoni, for monitoring quality requirements using architectural information, such as architectural styles. To assess the practical use of the proposed process, an evaluation experiment with students from the Computer Science course was performed. Students performed the monitoring of a real application called Falibras. This target system performs online text translation from Portuguese to the Brazilian Sign Language (Libras). Through the experiment, it was observed that the students considered the proposed process feasible to direct monitoring decisions. On the other hand, students who did not have access to the proposed process and do not know the architecture of Falibras have related more difficult to analyze the system. Thus, we have realized that it is important for the software developer to have guidelines that assist him/her in understanding the relevant quality requirements for the software that is being monitored, such as a processes that address the monitoring of these requirements. Preliminary results indicate that the ArMoni process contributes to guide the monitoring and analysis of the system based on architectural information, such as the architectural styles and we have noticed that it benefits mainly less experienced developers. Thus, this work presents academic relevance in software engineering and can assist developers and software engineers to monitor quality requirements, such as reliability and efficiency, as well as to identify possible structural bottlenecks that could compromise the satisfaction of these requirements.