Dateset of “Review of Software Performance and Architecture Analysis Integration”
<p><span class="fontstyle0">Software architecture is the most important determinant<br />to systematically achieve quality attributes in a software<br />system, including software performance. However there is no<br />comprehensive understanding of why and how to integrate<br />software architecture and performance analysis to guide research.<br />To fill this gap, this paper presents a systematic review of 83<br />studies. We focus on six research questions that provide guidance<br />for researchers and practitioners to gain an in-depth understanding<br />of this research area. These questions addressed: the study purposes<br />(RQ1), the relevant software development activities (RQ2), the<br />typical study templates (RQ3), the available instruments for<br />automating the analysis (RQ4), the evaluation methodology (RQ5),<br />and the limitations and future research directions (RQ6). The most<br />important outcome of our study is an identification of critical<br />research gaps and directions, including: 1) the lack of available tools<br />and benchmark datasets to support replication, cross-validation and<br />comparison of studies; 2) the need for architecture and performance<br />analysis techniques that handle the challenges with emerging<br />software domains, such as blockchain and cyber-physical systems;<br />3) the lack of consideration of factors such as usability that<br />impact the practical adoption of the architecture and performance<br />analysis approaches; 4) the need for integrating architecture and<br />performance analysis in Agile settings; and finally 5) the need for<br />the adoption of modern ML/AL techniques to efficiently integrate<br />architecture and performance analysis.</span></p>
The reviewed studies are listed in the attached spreadsheet.