INTELLIGENT METHOD FOR MONITORING AND RESOURCE OPTIMIZATION OF IT INFRASTRUCTURE BASED ON MACHINE LEARNING
DOI:
https://doi.org/10.34185/1991-7848.itmm.2026.01.091Keywords:
IT infrastructure, monitoring, machine learning, load forecasting, AIOps, resource balancing, neuro-fuzzy systemsAbstract
The paper analyzes modern approaches to monitoring and load management in IT infrastructures. The limitations of traditional reactive methods are considered and the expediency of moving to intelligent AIOps systems is substantiated. The authors proposed a concept of a monitoring method based on the use of machine learning algorithms (in particular, LSTM models and neuro-fuzzy networks) for proactive forecasting of load time series. Particular attention is paid to intelligent resource balancing in cloud and microservice environments, which allows minimizing latency and optimizing infrastructure costs. The results can be used in the development of adaptive autoscaling systems.
References
Designing Data-Intensive Applications / M. Kleppmann. — O’Reilly Media, 2017.
Observability Engineering / C. Majors et al. — O’Reilly, 2022.
AIOps: Real-World Use Cases and Platforms / S. Dang et al. — Springer, 2021.
Time Series Analysis: Forecasting and Control / G. Box, G. Jenkins. — Wiley, 2015.
Deep Learning / I. Goodfellow, Y. Bengio, A. Courville. — MIT Press, 2016.
Hands-On Machine Learning with Scikit-Learn, Keras, and TensorFlow / A. Géron. — O’Reilly, 2022.
Kubernetes: Up and Running / K. Burns, B. Grant. — O’Reilly, 2022.
Cloud Native Patterns / C. Richardson. — Manning, 2018.
An SLO-Driven and Cost-Aware Autoscaling Framework for Kubernetes / V. Punniyamoorthy, B. Kumar, S. Saha, L. Butra, M. Palanigounder, A. K. Agarwal, K. Kannan — arXiv, 2025. DOI: https://doi.org/10.48550/arXiv.2512.23415
