MACHINE LEARNING-BASED PERFORMANCE PREDICTION AND OPTIMIZATION IN MICROSERVICES-ORIENTED ARTIFICIAL INTELLIGENCE SYSTEMS

Guillaume Pierre Menaud

MACHINE LEARNING-BASED PERFORMANCE PREDICTION AND OPTIMIZATION IN MICROSERVICES-ORIENTED ARTIFICIAL INTELLIGENCE SYSTEMS

Authors

Guillaume Pierre Menaud

Professor and Researcher, Thailand.

Keywords:

Microservices, Machine Learning, Performance Prediction, Optimization, AI Systems, Reinforcement Learning, Telemetry

Synopsis

Purpose: This paper investigates the integration of machine learning (ML) techniques for enhancing the performance prediction and optimization of microservices-oriented AI systems, which are increasingly employed in modern software architectures.

Design/methodology/approach: We propose a hybrid ML-based framework that combines historical trace analysis and real-time telemetry data for performance forecasting and optimization. The architecture uses supervised learning for prediction and reinforcement learning for adaptive optimization.

Findings: Our review and proposed model demonstrate that ML significantly improves the adaptability and efficiency of microservices under varying workloads, reducing latency and increasing throughput by over 30% in simulated deployments.

Practical implications: The findings help engineers deploy scalable, fault-tolerant, and efficient microservices AI platforms in real-world scenarios such as edge computing, IoT, and cloud-based services.

Originality/value: This work provides a novel combination of predictive modeling and real-time optimization in microservices with ML, offering comparative insights and practical deployment considerations.

References

[1] Leng, X., Juang, T. H., Chen, Y., & Liu, H. (2019). AOMO: an AI-aided optimizer for microservices orchestration. ACM SIGCOMM. https://doi.org/10.1145/3342280.3342287

[2] Harun, H. (2019). AI-Based Optimization of Resource Utilization in Edge and Cloud Environments. AIJCST. https://aijcst.org/index.php/aijcst/article/view/13

[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] Zwietasch, T. (2017). Online Failure Prediction for Microservice Architectures. Stuttgart University. https://elib.uni-stuttgart.de/handle/11682/9290

[5] Kaul, D. (2019). Optimizing Resource Allocation in Multi-Cloud Environments with Artificial Intelligence. JICET.

[6] Thompson, A. (2019). Integrating Automation and AI in Microservices Architectures. IJAI. https://itaimle.com/index.php/ijaiml/article/view/80

[7] Felstaine, E., & Hermoni, O. (2018). Machine Learning, Containers, and Microservices. In AI for Autonomous Networks. https://www.taylorfrancis.com

[8] Magableh, B., & Almiani, M. (2019). Deep Q-Learning for Self-Adaptive Microservices. IEEE Access.

[9] Liu, B. (2019). Study and Benchmarking of AI Model Serving Systems. Aalto University. https://aaltodoc.aalto.fi

[10] Ortiz, G., Caravaca, J. A., & García-de-Prado, A. (2019). Context-Aware Microservices for Predictive Analytics. IEEE IoT. https://ieeexplore.ieee.org/document/8936407

[11] Zhou, X., Peng, X., Xie, T., Sun, J., Ji, C., & Liu, D. (2019). Latent Error Prediction in Microservices Using ML. ACM. https://doi.org/10.1145/3338906.3338961