# Sustainability in Software Architecture: A Systematic Mapping Study

**Authors:** Vasilios Andrikopoulos, Rares-Dorian Boza, Carlos Perales, Patricia Lago  
**arXiv:** [2204.11657](https://arxiv.org/abs/2204.11657)  
**Date:** 2022-04-25  
**Summary page:** https://www.summarizepaper.com/en/paper/2204.11657v1/

## Key points

- Lack of research specifically focused on software architecture and sustainability
- Conducted a systematic mapping study to reflect on the current state of the art and identify needs for further research
- Existing works have primarily focused on specific aspects of sustainability, neglecting the holistic perspective required
- More reflection-oriented research works and better coverage of activities in the architecting life cycle are needed to advance the maturity of this area
- Proposed a research agenda for sustainability-aware software architecture based on findings
- Study design followed guidelines for systematic mapping studies in software engineering by Petersen et al.
- Defined research questions addressed publication trends, coverage of architecting life cycle activities, and dimensions of sustainability as a software quality
- Data extraction performed manually alongside finalizing paper filtering due to insufficiencies in automated methods
- Snowballing used to locate additional relevant studies with separate data extraction for these studies
- Detailed discussion provided for each phase of the study design including actions taken during its conduct
- Summary of findings in Section III, synthesis into a research agenda in Section IV, threats to validity discussed in Section V, and conclusion with key points in Section VI

## Summary

This paper focuses on the intersection between sustainability and software architecture research. While there have been secondary studies addressing sustainability in software engineering, there has been a lack of research specifically focused on software architecture. To fill this gap, the authors conducted a systematic mapping study to reflect on the current state of the art and identify needs for further research. The findings of the study show that existing works have primarily focused on specific aspects of sustainability, particularly technical and "inward facing" ones. This narrow focus neglects the holistic perspective required to address sustainability as a multi-faceted concern. The authors argue that more reflection-oriented research works and better coverage of activities in the architecting life cycle are needed to advance the maturity of this area. Based on their findings, the authors propose a research agenda for sustainability-aware software architecture. The study design followed guidelines for systematic mapping studies in software engineering by Petersen et al. The process involved defining research questions to scope the study, searching and filtering relevant papers based on inclusion/exclusion criteria, classifying selected papers through keywording abstracts, extracting relevant data for mapping, and preparing a map. Some amendments were made to this process due to insufficiencies in keyword extraction from abstracts alone and difficulties in filtering papers solely based on automated methods. The defined research questions addressed publication trends in sustainability-related software architecture research, coverage of architecting life cycle activities when considering sustainability, and discussion of dimensions of sustainability as a software quality. These perspectives guided data collection and analysis. Data extraction was performed manually alongside finalizing paper filtering due to insufficiencies in automated methods. Snowballing was also used to locate additional relevant studies with separate data extraction for these studies. The paper provides a detailed discussion of each phase of the study design including actions taken during its conduct; Section III summarizes the findings to answer identified research questions; Section IV synthesizes these findings into a research agenda for sustainability-aware software architecture; Section V discusses threats to validity; and finally Section VI concludes with a summary of key points. Overall, this paper contributes to understanding of sustainability in software architecture research and highlights need for more comprehensive approach to address it as multi-faceted concern.