ADAPTIVE MICROSERVICES FOR MACHINE LEARNING PIPELINES: LEVERAGING ARTIFICIAL INTELLIGENCE FOR AUTOMATED SCALING AND FAULT TOLERANCE

Romero Felipe Clara

ADAPTIVE MICROSERVICES FOR MACHINE LEARNING PIPELINES: LEVERAGING ARTIFICIAL INTELLIGENCE FOR AUTOMATED SCALING AND FAULT TOLERANCE

Authors

Romero Felipe Clara

Independent Researcher, United Kingdom.

Keywords:

Microservices, Machine Learning Pipelines, Scalability, Fault Tolerance, Automation, Artificial Intelligence, Orchestration, Containerization

Synopsis

Purpose: This paper explores how adaptive microservice architectures can enhance machine learning (ML) pipelines by incorporating AI-driven mechanisms for automated scaling and fault tolerance.

Design/methodology/approach: We integrate insights from historical research into microservices and ML pipeline engineering, emphasizing decentralized container orchestration, AI-based monitoring, and automated failure recovery. Diagrams and comparative tables support this model.

Findings: The study reveals that AI-integrated microservices improve scalability and resilience in real-time systems. Automated feedback loops reduce manual interventions and improve uptime.

Practical Implications: AI-enhanced microservices can significantly reduce system downtime and operational overheads in ML production environments, especially where dynamic workloads are expected.

Originality/value: We present a synthesized, scalable, fault-tolerant design model for ML workflows that can autonomously manage resources, a novel contribution that merges AI, microservices, and pipeline orchestration strategies.

References

[1] Boag, S., Dube, P., Herta, B., & Hummer, W. (2017). Scalable multi-framework lifecycle management of deep learning training jobs. LearningSys@NIPS. http://learningsys.org/nips17/assets/papers/paper_29.pdf

[2] 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

[3] Devarakonda, R. R. (2017). A Microservices-Based Approach for Scalable Deployment of Machine Learning Models. SSRN. https://papers.ssrn.com/sol3/papers.cfm?abstract_id=5234707

[4] Felstaine, E., & Hermoni, O. (2018). Machine Learning, Containers, Cloud Natives, and Microservices. CRC Press. https://www.taylorfrancis.com/chapters/edit/10.1201/9781351130165-7

[5] Fowler, S. J. (2016). Production-ready microservices. O'Reilly.

[6] Heorhiadi, V., Rajagopalan, S., & Jamjoom, H. (2016). Gremlin: Systematic resilience testing of microservices. IEEE ICDCS. https://ieeexplore.ieee.org/document/7536505

[7] Khan, M. F. B. (2016). Spring Boot and Microservices. ResearchGate. https://www.researchgate.net/publication/386554934

[8] Khoonsari, P. E., et al. (2019). Interoperable and scalable data analysis with microservices. Bioinformatics, 35(19), 3752–3760. https://doi.org/10.1093/bioinformatics/btz379

[9] Rasi, N. (2018). TensorFlow Microservices with QoS Guarantees. Politecnico di Milano. https://www.politesi.polimi.it/handle/10589/152321

[10] Thompson, A. (2019). Integrating Automation and AI in Microservices Architectures. IJAIML. https://itaimle.com/index.php/ijaiml/article/download/80/203

[11] Timur, T. D., & Purnama, I. K. E. (2019). Deploying scalable face recognition pipeline using distributed microservices. IEEE Conference Proceedings. https://ieeexplore.ieee.org/document/8973287