Managing Model Versioning and Lifecycle in Artificial Intelligence Systems Using Microservices

Barbara Mintzes Elton

Managing Model Versioning and Lifecycle in Artificial Intelligence Systems Using Microservices

Authors

Barbara Mintzes Elton

AI Model Lifecycle & Platform Engineer , Spain

Keywords:

ModelOps, Microservices, AI Lifecycle, Model Versioning, DevOps, MLOps, Cloud-native, Containerization

Synopsis

Purpose: This paper investigates the integration of microservices to manage AI model versioning and lifecycle, focusing on scalability, reproducibility, and operational efficiency.

Design/methodology/approach: We reviewed the existing literature and frameworks on microservices and AI lifecycle management. A comparative analysis of architecture models and management strategies is presented alongside proposed enhancements.

Findings: ModelOps practices, when integrated with microservice-based architectures, offer superior flexibility in version control, reproducibility, and deployment automation. Cloud-native ecosystems further enhance dynamic lifecycle governance.

Practical implications: AI model governance, auditing, and rollback are critical in production environments. Microservices decouple lifecycle components, enabling agile experimentation and secure deployment.

Originality/value: This paper bridges the gap between AI lifecycle best practices and modern cloud-native deployment, offering a roadmap to scalable model management.

References

(1) Bhattacharjee, A., Barve, Y., Khare, S., et al. (2019). Stratum: A BigData-as-a-Service for Lifecycle Management of IoT Analytics Applications. IEEE Big Data. https://ieeexplore.ieee.org/document/9006518

(2) Ciucu, R., Adochiei, F. C., Adochiei, I. R., & Argatu, F. (2019). Innovative DevOps for Artificial Intelligence. Scientific Bulletin, 24(1), 65–71.

(3) Gummadi, V. P. K. (2019). Microservices architecture with APIs: Design, implementation, and MuleSoft integration. Journal of Electrical Systems, 15(4), 130–134. https://doi.org/10.52783/jes.9328

(4) Hummer, W., Muthusamy, V., & Rausch, T. (2019). ModelOps: Cloud-Based Lifecycle Management for Reliable and Trusted AI. IEEE International Conference on Cloud Engineering. https://ieeexplore.ieee.org/document/8790192

(5) Horn, G., Arrieta, L. O. E., Di Martino, B., & Skrzypek, P. (2019). Dynamic Patterns for Cloud Application Life-Cycle Management. Springer Conference on P2P Computing.

(6) Alvarez-Rodríguez, J. M., & Zuñiga, R. M. (2019). Integration of Systems Engineering and the AI/ML Model Lifecycle. INCOSE International Symposium, 29(1), 1335–1352.

(7) Rao, D. J. (2019). AI in the Modern Software World. Wiley.

(8) Santhanam, P., Farchi, E., & Pankratius, V. (2019). Engineering Reliable Deep Learning Systems. arXiv preprint arXiv:1910.12582.

(9) Spell, D. C., Zeng, X. H. T., & Chung, J. Y. (2017). Flux: Groupon’s Automated, Scalable ML Platform. IEEE Big Data.

(10) Lipton, P., Palma, D., & Rutkowski, M. (2018). TOSCA Solves Big Problems in the Cloud and Beyond. IEEE Cloud Computing, 5(1), 46–54.

(11) O'Connor, R. V., Elger, P., & Clarke, P. M. (2017). Continuous Software Engineering—A Microservices Architecture Perspective. Journal of Software: Evolution and Process, 29(11), e1866.