Carregando...
 

Notí­cias

Pesquisador do GSORT publica artigo no SBES 2014

Sábado, 21 de Junho de 2014

O Simpósio Brasileiro de Engenharia de Software (SBES) é o mais tradicional simpósio brasileiro da SBC (Sociedade Brasileira de Computação) na área de desenvolvimento de software, indo para sua 28ª edição no ano de 2014.

O SBES se consolidou como o mais importante evento técnico-científico na área de engenharia de software no Brasil, atraindo diversos participantes entre pesquisadores, estudantes e profissionais.

É com prazer que informamos a aceitação do artigo "Do Search-Based Approaches Improve the Design of Self-Adaptive Systems ? A Controlled Experiment", de autoria do Prof. Sandro Andrade (GSORT/IFBa) e Prof. Raimundo Macêdo (LaSiD/UFBa). Mais informações abaixo. O evento acontece de 28 de Setembro a 3 de Outubro de 2014, em Maceió (AL).

Parabéns aos autores !

Título: Do Search-Based Approaches Improve the Design of Self-Adaptive Systems ? A Controlled Experiment

Autores: Sandro S. Andrade e Raimundo J. de A. Macêdo

Resumo: Endowing software systems with self-adaptation capabilities has shown to be quite effective in coping with uncertain and dynamic operational environments as well as managing the complexity generated by non-functional requirements. Nowadays, a large number of approaches tackle the issue of enabling self-adaptive behavior from different perspectives and under diverse assumptions, making it harder for architects to make judicious decisions about design alternatives and quality attributes trade-offs. It has currently been claimed that search-based software design approaches may improve the quality of resulting artifacts and the productivity of design processes, as a consequence of promoting a more comprehensive and systematic design knowledge representation and preventing design bias and false intuition. To the best of our knowledge, no controlled experiments have been performed to provide sound evidence of such claim in the self-adaptive systems domain. In this paper, we report the results of a controlled experiment performed with 24 students of a graduate program in Distributed and Ubiquitous Computing. The experiment evaluated the design of self-adaptive systems using a search-based approach, in contrast to the use of a style-based non-automated approach. The results show that search-based approaches can improve the effectiveness of resulting architectures and reduce design complexity. We found no evidence regarding the method's potential for leveraging the acquisition of distilled design knowledge by novice software architects.