A 2024 survey of 53 industry practitioners (Nogueira et al., ICSME 2024) documented how monolith-to-microservice migrations are actually conducted and found that the operational practices on which migration's promised benefits depend remain ad hoc: testing relies on unit and integration tests because extensive end-to-end system tests are impractical at high release frequency and about half of the systems are polyglot, while database management splits between centralized and decentralized databases with no evaluated strategy for data consistency or for partitioning legacy schemas in the presence of legacy, BI, compliance, and cost barriers. The authors explicitly call for future research on testing methodologies and database management in microservice architectures. The gap remains open: 2025 systematic reviews of microservices testing map the challenge space but leave solution gaps; the authors' own 2026 industrial case study addresses only decomposition and confirms the data-side neglect; and no controlled evaluation of eventual-consistency, CQRS, or event-sourcing strategies under realistic migration constraints exists.

Background

In a microservice architecture, an application is decomposed into independently deployable services that communicate over a network, each typically owning its own data, in contrast to a monolith in which all functionality runs as a single deployable unit sharing one database. Migrating a legacy monolith to microservices is a widespread industrial modernization path, motivated mainly by desires for easier maintenance, better scalability, and faster deployment. The reference point for this problem is a mixed-methods survey by Nogueira, Felizardo, Amaral, Assunção, and Colanzi, presented at the IEEE International Conference on Software Maintenance and Evolution (ICSME) 2024, of 53 practitioners across diverse countries, company sizes, and domains, which examined the motivations, activities, data-consistency strategies, and challenges of such migrations. Its findings define the practice baseline for this problem: unit testing (83% of participants) and integration testing (79%) dominate, because extensive end-to-end (that is, full-system) tests are impractical given the higher frequency of releases, a constraint already flagged by Heinrich et al. at the ACM/SPEC International Conference on Performance Engineering (ICPE 2017); 47% of participants use more than one programming language (polyglot stacks), which complicates instrumentation and monitoring across heterogeneous technology stacks; 58% use a dedicated database per service, while consistency is managed by eventual consistency controlled by the application (42%) or distributed/atomic transactions (30%), with CQRS (Command Query Responsibility Segregation, a pattern separating write and read models) and event sourcing (storing state as a sequence of change events) appearing as practitioner strategies. Barriers to partitioning a shared legacy database include proprietary and legacy dependencies — one participant's database was shared with the business-intelligence (BI) system, i.e., the analytical reporting consumers of operational data — licensing costs, context-separation complexity, and data-compliance and integrity requirements. Duplicated code across services (reported by 28% of participants) is handled by shared libraries or service calls, but without traceability or centralized management. Earlier literature had already flagged these areas: Heinrich et al. (ICPE 2017) listed performance-engineering research challenges including testing under frequent releases, Ben Ali (2019) identified microservice data-management strategies as an open challenge area, and Zhou et al. in the Journal of Systems and Software (JSS, 2023) documented practices and pains of microservice adoption in industry. What no prior or later work supplies is evaluated, migration-specific knowledge: which consistency and partitioning strategies actually work under realistic migration constraints, and how to test a system during migration when full end-to-end testing is impractical. The survey's authors state this explicitly: the study 'lays the groundwork for future research, particularly in areas such as testing methodologies and database management in microservice architectures', and its conclusion points to 'the need for continued research and tailored solutions to support successful microservice adoption'.

Problem Statement

For monolith-to-microservice migrations as practiced in industry — characterized by the 2024 ICSME practitioner survey — establish the missing methodological and empirical basis for the two migration-critical operational practice areas that the survey found unresolved and its authors named as the target of future research, and determine through that evidence whether the claimed benefits of migration (maintainability, scalability, faster deployment) are operationally realizable: (1) develop and empirically assess testing methodologies and toolchains tailored to migration settings, i.e., supporting validation of a partially migrated system when extensive end-to-end tests are impractical because of high release frequency and when the stack is polyglot, evaluated in realistic migration contexts rather than only on toy systems; and (2) produce controlled evaluations covering all four of the database-partitioning and data-consistency strategies practitioners actually report (eventual consistency controlled by the application, distributed transactions, CQRS, event sourcing) under realistic migration operational constraints, accounting for the documented barriers (shared legacy and BI consumers, data-compliance and integrity requirements, licensing cost, context-separation complexity), and including the transformation of a legacy data model into decentralized per-service schemas while preserving compliance and integrity for legacy consumers. These two components are jointly posed as a single objective — replacing the documented ad hoc practice with evaluated, tailored solutions for migration — and the verdict on operational realizability (maintainability, scalability, faster deployment) follows from the evidence of both components rather than from either alone.

References

  1. Vinicius L. Nogueira, Fernando S. Felizardo, Aline M. M. M. Amaral, Wesley K. G. Assunção, Thelma E. Colanzi (2024), 'Insights on Microservice Architecture Through the Eyes of Industry Practitioners', 2024 IEEE International Conference on Software Maintenance and Evolution (ICSME), pp. 765-777, DOI: 10.1109/ICSME58944.2024.00080, arXiv:2408.10434, https://arxiv.org/abs/2408.10434
  2. Fernando S. Felizardo, Vinicius L. Nogueira, João S. Castanho, Thelma E. Colanzi, Aline M. M. M. Amaral, Wesley K. G. Assunção (2026), 'Microservices by Optimization and Experience: Lessons from a Large-Scale Industrial Migration', 2026 IEEE International Conference on Software Analysis, Evolution and Reengineering (SANER), pp. 503-514, DOI: 10.1109/SANER67736.2026.00061, https://doi.org/10.1109/SANER67736.2026.00061
  3. Mingxuan Hui, Lu Wang, Hao Li, Ren Yang, Yuxin Song, Huiying Zhuang, Di Cui, Qingshan Li (2025), 'Unveiling the microservices testing methods, challenges, solutions, and solutions gaps: A systematic mapping study', Journal of Systems and Software, DOI: 10.1016/j.jss.2024.112232, https://doi.org/10.1016/j.jss.2024.112232
  4. Francisco Ponce, Roberto Verdecchia, Breno Miranda, Jacopo Soldani (2025), 'Microservices testing: A systematic literature review', Information and Software Technology, DOI: 10.1016/j.infsof.2025.107870, https://doi.org/10.1016/j.infsof.2025.107870
  5. Lena Gregor (2025), 'On Service-to-Service Integration Testing in Microservice Systems', 2025 IEEE Conference on Software Testing, Verification and Validation (ICST), pp. 794-796, DOI: 10.1109/ICST62969.2025.10988961, https://doi.org/10.1109/ICST62969.2025.10988961
  6. Hang-Wei Yeh, Shang-Pin Ma, Yi Chen (2024), 'Test Case Migration from Monolith to Microservices Using Large Language Models', 2024 IEEE International Conference on e-Business Engineering (ICEBE), pp. 29-35, DOI: 10.1109/ICEBE62490.2024.00014, https://doi.org/10.1109/ICEBE62490.2024.00014
  7. Georg-Daniel Schwarz, Felix Quast, Dirk Riehle (2025), 'Ensuring Syntactic Interoperability Using Consumer-Driven Contract Testing', Software Testing, Verification and Reliability, DOI: 10.1002/stvr.70006, https://doi.org/10.1002/stvr.70006
  8. Willem Meijer, Catia Trubiani, Aldeida Aleti (2024), 'Experimental evaluation of architectural software performance design patterns in microservices', Journal of Systems and Software, DOI: 10.1016/j.jss.2024.112183, https://doi.org/10.1016/j.jss.2024.112183
  9. Deepika Badampudi, Muhammad Usman, Xingru Chen (2024), 'Large scale reuse of microservices using CI/CD and InnerSource practices - a case study', Empirical Software Engineering, DOI: 10.1007/s10664-024-10595-w, https://doi.org/10.1007/s10664-024-10595-w
  10. Amr S. Abdelfattah, Tomás Cerny, Marwa Elsayed (2025), 'Empirical Insights into Microservice Language Heterogeneity in Practice', 2025 ACM/IEEE International Symposium on Empirical Software Engineering and Measurement (ESEM), pp. 458-466, DOI: 10.1109/ESEM64174.2025.00060, https://doi.org/10.1109/ESEM64174.2025.00060
  11. Xin Zhou, Shanshan Li, Lingli Cao, He Zhang, Zijia Jia, Chenxing Zhong, Zhihao Shan, Muhammad Ali Babar (2023), 'Revisiting the practices and pains of microservice architecture in reality: An industrial inquiry', Journal of Systems and Software 195:111521, DOI: 10.1016/j.jss.2022.111521, https://doi.org/10.1016/j.jss.2022.111521
  12. Robert Heinrich, André van Hoorn, Holger Knoche, Fei Li, Lucy Ellen Lwakatare, Claus Pahl, Stefan Schulte, Johannes Wettinger (2017), 'Performance Engineering for Microservices: Research Challenges and Directions', Proceedings of the 8th ACM/SPEC International Conference on Performance Engineering Companion, pp. 223-226, DOI: 10.1145/3053600.3053653, https://doi.org/10.1145/3053600.3053653
  13. Ahmed Ben Ali (2019), 'Data Management Strategies for Microservices: Challenges and Future Research Directions', International Journal of Artificial Intelligence and Digital Transformation, DOI: 10.67228/30713315/ijaidt-2019pi8q1d, https://doi.org/10.67228/30713315/ijaidt-2019pi8q1d
  14. Vinicius Lopes Nogueira (2024), 'Microservice migration survey and responses', Zenodo dataset, DOI: 10.5281/zenodo.13324295, https://doi.org/10.5281/zenodo.13324295, https://zenodo.org/records/13324295