Analysis of Microservice Coupling Measures
Rahman, Mohammad Imranur (2022)
2022
Master's Programme in Information Technology
Informaatioteknologian ja viestinnän tiedekunta - Faculty of Information Technology and Communication Sciences
This publication is copyrighted. You may download, display and print it for Your own personal use. Commercial use is prohibited.
Hyväksymispäivämäärä
2022-05-19
Julkaisun pysyvä osoite on
https://urn.fi/URN:NBN:fi:tuni-202204273988
https://urn.fi/URN:NBN:fi:tuni-202204273988
Tiivistelmä
Microservices architectures are composed of a collection of modular, fault-tolerant services. In recent years, the software engineering community has published research on viable, recurring, and effective architectural patterns in microservices-based architectures, as they are critical to the maintenance and scaling of microservice-based systems. As well as, ensuring low coupling and strong cohesion among the microservices that comprise the cloud-native application is a crucial property.Services that are loosely connected and highly coherent allow development teams to work in parallel, eliminating communication overhead between teams.
In the first section of this thesis, we attempted to generate a dataset by starting with a selected list of microservice-based projects. The collection is made up of 20 open-source applications that all use certain microservice architecture patterns. Furthermore, the dataset includes information about the aforementioned projects’ interservice calls and dependencies.
In the second section, we suggested methods for computing and visualizing the coupling be- tween microservices by expanding and adapting the notions underlying standard of structural coupling calculation. We validate these measures using a case study of 17 projects selected from the aforementioned dataset, and we propose an automated method for measuring them. The findings of this study emphasize how these metrics give practitioners with quantitative and visual views of service architecture, that can be used to design advanced measures to monitor the development of services.
In the first section of this thesis, we attempted to generate a dataset by starting with a selected list of microservice-based projects. The collection is made up of 20 open-source applications that all use certain microservice architecture patterns. Furthermore, the dataset includes information about the aforementioned projects’ interservice calls and dependencies.
In the second section, we suggested methods for computing and visualizing the coupling be- tween microservices by expanding and adapting the notions underlying standard of structural coupling calculation. We validate these measures using a case study of 17 projects selected from the aforementioned dataset, and we propose an automated method for measuring them. The findings of this study emphasize how these metrics give practitioners with quantitative and visual views of service architecture, that can be used to design advanced measures to monitor the development of services.