Métricas em microsserviços: análise da ISO/IEC 25010 e recomendações por LLM

Abstract

Introduction: The microservices pattern, which organizes a system into a collection of independent services, promotes flexibility in development, accelerating the delivery cycle and improving software quality. This model facilitates adaptation to changes, reducing development time and increasing team satisfaction. Objective: This study aims to collect performance and quality metrics from academia through a theoretical and systematic review and to understand the industry’s reality through an opinion survey. The goal is to identify challenges and solutions for managing microservice teams, relating quality metrics to the ISO/IEC 25010 Standard requirements, advancing scientific research, and supporting the management of microservice-based technologies with adaptable metrics that integrate artificial intelligence. Methodology: The Design Science Research method was adopted, where, in the case study, two artifacts were developed: a metrics catalog and a metrics recommendation tool based on this catalog to support microservice management. Conclusion: This work contributes to understanding academic and industrial needs by comparing both areas and proposing metrics to guide researchers, managers, and developers. Furthermore, it provides data-driven insights for decision-making in microservice implementation and is a foundation for future research. Future studies will focus on advancing Large Language Models (LLM), user engagement in LLM-based tools, and improving the LLM developed in this study, potentially impacting academia and industry, optimizing microservice management, and enhancing its quality.