Straddling the crevasse: A review of microservice software architecture foundations and recent advancements

Abstract

Microservice architecture style has been gaining wide impetus in the software engineering industry. Researchers and practitioners have adopted the microservices concepts into several application domains such as the internet of things, cloud computing, service computing, and healthcare. Applications developed in alignment with the microservices principles require an underlying platform with management capabilities to coordinate the different microservice units and ensure that the application functionalities are delivered to the user. A multitude of approaches has been proposed for the various tasks in microservices-based systems. However, since the field is relatively young, there is a need to organize the different research works. In this study, we present a comprehensive review of the research approaches directed toward microservice architectures and propose a multilevel taxonomy to categorize the existing research. The study also discusses the different distributed computing paradigms employing microservices and identifies the open research challenges in the domain.

Microservice transition and its granularity problem: A systematic mapping study

“…However, it does not focus on microservice granularity so our study is significant to understanding this microservitization challenge. In Reference 199, a broad approach is taken to study microservice literature. We acknowledge that it goes beyond our work in studying the research approaches utilised in the microservice community.…”

“…Furthermore, this study dives into the computing aspect of the transition to microservices (analogous to the development publications category in our study). Nevertheless, our study is goes beyond 199 with regards to defining the transition to microservices. Furthermore, our work is unique with regards to studying literature along different aspects of microservice granularity.…”

KubCG: A dynamic Kubernetes scheduler for heterogeneous clusters

“…A recent publication by Rodriguez et al 21 identifies the need for cloud-aware placement algorithms to consider the heterogeneities of the underlying resources (including parameters such as resource types and sizes). Joseph et al 22 state that resource request profiling and estimation is a research challenge that will aid the allocation and scheduling processes in the microservice systems.…”

Component‐aware orchestration of cloud‐based enterprise applications, from TOSCA to Docker and Kubernetes

“…9 Container orchestrators are then used to automate the deployment and management of containerised applications at a large scale. 10 Docker Swarm (https://docs.docker.com/engine/swarm) and Kubernetes (https://kubernetes. io) are currently the most popular solutions for orchestrating Docker containers, providing all necessary abstractions for scaling, discovering, load-balancing and interconnecting Docker containers over single-and multi-host systems.…”

“…Docker (https://www.docker.com) constitutes the de‐facto standard for container‐based virtualization, and it permits packaging software components (together with all software dependencies they need to run) in Docker images, which are then exploited as read‐only templates to create and run Docker containers 9 . Container orchestrators are then used to automate the deployment and management of containerised applications at a large scale 10 . Docker Swarm (https://docs.docker.com/engine/swarm) and Kubernetes (https://kubernetes.io) are currently the most popular solutions for orchestrating Docker containers, providing all necessary abstractions for scaling, discovering, load‐balancing and interconnecting Docker containers over single‐ and multi‐host systems 11,12…”