From legacy to microservices: A type‐based approach for microservices identification using machine learning and semantic analysis

Abstract

The microservices architecture (MSA) style has been gaining interest in recent years because of its high scalability, ability to be deployed in the cloud, and suitability for DevOps practices. While new applications can adopt MSA from their inception, many legacy monolithic systems must be migrated to an MSA to benefit from the advantages of this architectural style. To support the migration process, we propose MicroMiner, a microservices identification approach that is based on static‐relationship analyses between code elements as well as semantic analyses of the source code. Our approach relies on machine learning (ML) techniques and uses service types to guide the identification of microservices from legacy monolithic systems. We evaluate the efficiency of our approach on four systems and compare our results to ground‐truths and to those of two state‐of‐the‐art approaches. We perform a qualitative evaluation of the resulted microservices by analyzing the business capabilities of the identified microservices. Also a quantitative analysis using the state‐of‐the‐art metrics on independence of functionality and modularity of services was conducted. Our results show the effectiveness of our approach to automate one of the most time‐consuming steps in the migration of legacy systems to microservices. The proposed approach identifies architecturally significant microservices with a 68.15% precision and 77% recall.

Migration of Monolithic Systems to Microservices using AI: A Systematic Mapping Study

“…Louvain [Brito et al 2021, Li et al 2022, Fast Community [Eski and Buzluca 2018], Hierarchical K-means [Ren et al 2018], Leiden [Cao and Zhang 2022], SArF [Kamimura et al 2018] and SLINK [Kalia et al 2021] • Partitioning-based: Fuzzy C-means [Trabelsi et al 2022] and K-means [Bajaj et al 2020] • Unclassified: [Nunes et al 2019 Considering unsupervised techniques, the NSGA-II genetic algorithm is employed by the two studies in the category [Liu et al 2022, Jin et al 2021, while topic detection algorithms such LDA [Pigazzini et al 2019, Brito et al 2021 or SLDA [Pigazzini et al 2019] Kalia et al 2021, Gysel et al 2016, Matias et al 2020, Ren et al 2018, Sellami et al 2022, Brito et al 2021, Li et al 2022, Bajaj et al 2020, Trabelsi et al 2022 In the Analysis and Design (A&D) phases of an application, documentation is the principal output. However, several types of documentation might be used to migrate a monolithic application towards microservices, namely use cases [Gysel et al 2016], design documents [Li et al 2022, Gysel et al 2016, and Entity-Relationship design models [Li et al 2022, Gysel et al 2016.…”

“…Studies Percentage Static [Brito et al 2021, Kamimura et al 2018, Mathai et al 2022, Nitin et al 2022, Nunes et al 2019, Pigazzini et al 2019, Sellami et al 2022, Trabelsi et al 2022 36.4% Dynamic [Bajaj et al 2020, Jin et al 2021, Kalia et al 2021 [Eski andBuzluca 2018, Mazlami et al 2017] 9.1% Documentation (A&D) + Dynamic + Static [Li et al 2022] 4.5% Documentation based (A&D) [Gysel et al 2016] 4.5% By language support Although the majority of the reviewed techniques are designed to support a monolith developed in any programming language (65.2%), a subset of them (53.3%) currently support Java monoliths due to the fact those techniques rely on tools that are only available for the Java language, for instance Wala [Ren et al 2018] and Java Call Graph [Cao and Zhang 2022]. 21.7% of the studies migrate Java monoliths to microservices, thus becoming Java the most popular monolith language to develop migration techniques.…”

A systematic mapping study: The new age of software architecture from monolithic to microservice architecture—awareness and challenges

“…A fresh criterion, service power, is proposed to confirm the scaling of required services. MicroMiner is a type-based method for locating microservices in monolithic software systems, which was introduced by the author [4]. MicroMiner is guided by a taxonomy of service types that are anticipated using machine learning classification models.…”

A Comprehensive Microservice Extraction Approach Integrating Business Functions and Database Entities

“…Each service is accountable for one or more closelyrelated functions. However, finding these optimized sets of services is intellectually hard and takes time to implement as well [28]. Software architects who remodularize their monolithic applications to get microservices without clearly understanding its pros and cons invite risk and unforeseen problems [31].…”

“…Kumar et al [18] suggest moving a method to the class that uses it the most. This Move Method Refactoring (MMR) makes classes more cohesive internally and also eliminates dependency between classes [28]. Our research is motivated to find methods that are in need of refactoring.…”