This paper explores the application of the Isolation Forest algorithm for detecting anomalies in performance monitoring data of a SaaS project’s servers. The main hypothesis suggests that the algorithm can identify early signs of performance degradation and potential failures by analyzing basic metrics such as CPU load, memory usage, network traffic, and disk space. Two approaches were tested: analyzing each metric separately and aggregating them into a single indicator to assess the overall system state. The results showed that Isolation Forest demonstrates high sensitivity to sudden changes in metrics, leading to a significant number of false positives. This issue is particularly relevant when dealing with short-term metric spikes that do not necessarily indicate real system problems. The paper discusses the limitations of this approach, including the need for fine-tuning hyperparameters, and suggests possible solutions for improving anomaly detection accuracy, such as preprocessing data and combining it with other methods. This study highlights the importance of advanced machine learning techniques in server performance monitoring, especially in conditions with limited metrics, typical of closed-source systems.

KEREEV A.K.

PhD, Associate Professor of the Department of Computer Science and Information Technology, Aktobe Regional University named after K. Zhubanov, Aktobe, Kazakhstan

E-mail: akereyev@zhubanov.edu.kz, https://orcid.org/0000-0002-8283-5807

MIKHELSON O.Yu.

Senior Infrastructure Engineer, ActivSoft, Aktobe, Kazakhstan

E-mail: miol@activsoft.kz, https://orcid.org/0009-0009-6753-3120

1. Ronchieri E. Anomaly Detection in Data Center IT & Physical Infrastructure / Elisabetta Ronchieri, Luca Giommi, Luigi Benedettto Scarponi, Luca Torzi, Alessandro Costantini, Doina Cristina Duma, Davide Salomoni // EPJ Web of Conf. 295 07004 (2024) DOI: 10.1051/epjconf/202429507004 DOI: https://doi.org/10.1051/epjconf/202429507004
2. Bursic S. Anomaly Detection from Log Files Using Unsupervised Deep Learning. / Bursic Sathya, Cuculo Vittorio, D'Amelio Alessandro // Formal Methods. FM 2019 International Workshops. FM 2019. Lecture Notes in Computer Science (), vol 12232. Springer, Cham. DOI: 10.1007/978-3-030-54994-7_15 DOI: https://doi.org/10.1007/978-3-030-54994-7_15
3. Gu W. Performance Issue Identification in Cloud Systems with Relational-Temporal Anomaly Detection / Wenwei Gu, Jinyang Liu, Zhuangbin Chen, Jianping Zhang, Yuxin Su, Jiazhen Gu, Cong Feng, Zengyin Yang, Michael Lyu // arXiv, 2023. https://arxiv.org/abs/2307.10869
4. Liu F. "Isolation Forest" / Fei Tony Liu, Kai Ming Ting, Zhi-Hua Zhou // 2008 Eighth IEEE International Conference on Data Mining, Pisa, Italy, 2008, pp. 413-422, DOI: 10.1109/ICDM.2008.17. DOI: https://doi.org/10.1109/ICDM.2008.17
5. Prometheus. Prometheus documentation. [Электронный ресурс]. – Режим доступа: https://prometheus.io/docs/ (дата обращения: 07.10.2024).
6. Scikit Learn. Scikit Learn user guide. [Электронный ресурс]. – Режим доступа: https://scikit-learn.org/stable/user_guide.html (дата обращения: 07.10.2024).