Fostering Microservice Maintainability Assurance through a Comprehensive Framework

TL;DR

This paper presents an automated framework to assess and improve the maintainability of microservice systems by integrating static and dynamic analysis, visualization, and automated evaluations, validated through practitioner studies.

Contribution

It introduces a comprehensive, automated assessment framework specifically designed for microservice architectures, addressing maintainability challenges with multi-perspective analysis tools.

Findings

Framework effectively identifies maintainability issues in microservices.

Practitioner validation confirms improved understanding of system qualities.

Enhanced detection of coupling, testing coverage, and semantic clones.

Abstract

Cloud-native systems represent a significant leap in constructing scalable, large systems, employing microservice architecture as a key element in developing distributed systems through self-contained components. However, the decentralized nature of these systems, characterized by separate source codes and deployments, introduces challenges in assessing system qualities. Microservice-based systems, with their inherent complexity and the need for coordinated changes across multiple microservices, lack established best practices and guidelines, leading to difficulties in constructing and comprehending the holistic system view. This gap can result in performance degradation and increased maintenance costs, potentially requiring system refactoring. The main goal of this project is to offer maintainability assurance for microservice practitioners. It introduces an automated assessment framework tailored to microservice architecture, enhancing practitioners' understanding and analytical capabilities of the multiple system perspectives. The framework addresses various granularity levels, from artifacts to constructing holistic views of static and dynamic system characteristics. It integrates diverse perspectives, encompassing human-centric elements like architectural visualization and automated evaluations, including coupling detection, testing coverage measurement, and semantic clone identification. Validation studies involving practitioners demonstrate the framework's effectiveness in addressing diverse quality and maintainability issues, revealing insights not apparent when analyzing individual microservices in isolation.