koz"Вестник Северо-Казахстанского университета имени Манаша Козыбаева"Bulletin of Manash Kozybayev North Kazakhstan University2958-003X2958-0048М. Қозыбаев атындағы СҚУ10.54596/2958-0048-2026-2-266-277koz-2787Research ArticleИНФОРМАЦИОННО-КОММУНИКАЦИОННЫЕ ТЕХНОЛОГИИINFORMATION AND COMMUNICATION TECHNOLOGIESГРАФИЧЕСКИЕ НЕЙРОННЫЕ СЕТИ КАК ИНСТРУМЕНТ МОНИТОРИНГА И ОБНАРУЖЕНИЯ АНОМАЛИЙ В ОБЛАЧНЫХ СИСТЕМАХGRAPH NEURAL NETWORKS AS A TOOL FOR MONITORING AND DETECTING ANOMALIES IN CLOUD SYSTEMSAbdullahА.AbdullahA.

Серданг

Associate Professor, Faculty of Computer Science and Information Technology, Head of reseacrh group - wireless, mobile and quantum computing (WiMoC) 

Serdang

 

,

azizol@upm.edu.myФакультет компьютерных наук и информационных технологий, Университет Путра МалайзияМалайзияFaculty of Computer Science and Information Technolog Universiti Putra MalaysiaMalaysia20260907202602 (70)266277Copyright © Abdullah А., 20262026Abdullah А.Abdullah A.Данная работа распространяется под лицензией Creative Commons Attribution 4.0.This work is licensed under a Creative Commons Attribution 4.0 License.https://vestnik.ku.edu.kz/jour/article/view/2787

Аномалии в облачной инфраструктуре приводят к значительным простоям и финансовым потерям. Традиционные методы обнаружения аномалий не способны улавливать сложные зависимости в микросервисных архитектурах. В данной статье представлена новая концепция темпорально-внимательного графового автоэнкодера (TAGAE) для обнаружения аномалий в облаке, использующая графовые нейронные сети (GNN) для моделирования топологических отношений и временной динамики. Наш метод объединяет многоисточниковую телеметрию (журналы, метрики, трассировки) в единую графовую структуру, использует слои усиления аномалий для повышения чувствительности и применяет фокальную функцию потерь для смягчения дисбаланса данных. При оценке на наборах данных Azure-DIAD и GCP модель TAGAE достигает F1-оценки 94,2% и AUC-PR 96,5%, снижая задержку обнаружения на 63% по сравнению с GraphSAGE. Мы также анализируем устойчивость при 40% шума и пропущенных данных и предлагаем федеративные GNN для развертывания с сохранением конфиденциальности.

Cloud infrastructure anomalies cause significant downtime and financial losses. Traditional anomaly detection methods fail to capture complex dependencies in microservice architectures. This paper presents a novel Temporal-Attentive Graph Autoencoder (TAGAE) framework for cloud anomaly detection, leveraging Graph Neural Networks (GNNs) to model topological relationships and temporal dynamics. Our method integrates multi-source telemetry (logs, metrics, traces) into a unified graph structure, utilizes anomaly amplification layers for enhanced sensitivity, and employs focal loss for data imbalance mitigation. Evaluated on Azure-DIAD and GCP datasets, TAGAE achieves 94.2% F1-score and 96.5% AUC-PR, reducing detection latency by 63% compared to GraphSAGE. We further analyze robustness under 40% noise/missing data and propose federated GNNs for privacy-preserving deployment.

Графовые нейронные сетиобнаружение аномалийоблачная инфраструктуравременная графовая сверткаграфовый автоэнкодерзависимости микросервисовслияние телеметрииGraph Neural NetworksAnomaly DetectionCloud InfrastructureTemporal Graph ConvolutionGraph AutoencoderMicroservice DependenciesTelemetry FusionReferencesGarg, S., Kaur, K., Kumar, N., & Rodrigues, J. J. (2020). Ensembled Machine Learning-Based Anomaly Detection in Cloud Environment. ACM Transactions on Internet Technology. https://doi.org/10.1145/3377754Garg, S., Kaur, K., Kumar, N., & Rodrigues, J. J. (2020). Ensembled Machine Learning-Based Anomaly Detection in Cloud Environment. ACM Transactions on Internet Technology. https://doi.org/10.1145/3377754Ma, X., Wu, J., Xue, S., Yang, J., Zhou, C., Sheng, Q. Z., ... & Reynolds, L. (2021). A Comprehensive Survey on Graph Anomaly Detection with Deep Learning. IEEE Transactions on Knowledge and Data Engineering. https://doi.org/10.1109/TKDE.2021.3118815Ma, X., Wu, J., Xue, S., Yang, J., Zhou, C., Sheng, Q. Z., ... & Reynolds, L. (2021). A Comprehensive Survey on Graph Anomaly Detection with Deep Learning. IEEE Transactions on Knowledge and Data Engineering. https://doi.org/10.1109/TKDE.2021.3118815Wu, Z., Pan, S., Chen, F., Long, G., Zhang, C., & Yu, P. S. (2020). A Comprehensive Survey on Graph Neural Networks. IEEE Transactions on Neural Networks and Learning Systems. https://doi.org/10.1109/TNNLS.2020.2978386Wu, Z., Pan, S., Chen, F., Long, G., Zhang, C., & Yu, P. S. (2020). A Comprehensive Survey on Graph Neural Networks. IEEE Transactions on Neural Networks and Learning Systems. https://doi.org/10.1109/TNNLS.2020.2978386He, H., Li, X., Chen, P., Chen, J., Liu, M., & Wu, L. (2024). Efficiently localizing system anomalies for cloud infrastructures: A novel dynamic graph transformer based parallel framework. Journal of Cloud Computing, 13(115). https://doi.org/10.1186/s13677-024-00677-x.He, H., Li, X., Chen, P., Chen, J., Liu, M., & Wu, L. (2024). Efficiently localizing system anomalies for cloud infrastructures: A novel dynamic graph transformer based parallel framework. Journal of Cloud Computing, 13(115). https://doi.org/10.1186/s13677-024-00677-x.Zhang, Y., Li, Z., Zheng, Z., Chen, P., & Ma, D. (2022). CloudRCA: A Root Cause Analysis Framework for Cloud Computing Platforms. Proceedings of the 31st ACM International Conference on Information & Knowledge Management. https://doi.org/10.1145/3511808.3557121Zhang, Y., Li, Z., Zheng, Z., Chen, P., & Ma, D. (2022). CloudRCA: A Root Cause Analysis Framework for Cloud Computing Platforms. Proceedings of the 31st ACM International Conference on Information & Knowledge Management. https://doi.org/10.1145/3511808.3557121Wang, C., Peng, X., Yu, M., Li, P., Bao, T., & Zhao, J. (2021). MicroHECL: High-Efficient Root Cause Localization in Large-Scale Microservice Systems. IEEE/ACM 43rd International Conference on Software Engineering (ICSE). https://doi.org/10.1109/ICSE43902.2021.00040Wang, C., Peng, X., Yu, M., Li, P., Bao, T., & Zhao, J. (2021). MicroHECL: High-Efficient Root Cause Localization in Large-Scale Microservice Systems. IEEE/ACM 43rd International Conference on Software Engineering (ICSE). https://doi.org/10.1109/ICSE43902.2021.00040Zhou, X., Peng, X., Xie, T., Sun, J., Ji, C., Li, W., & Ding, D. (2021). Fault Anomaly Detection for Microservice Architecture Based on Graph Neural Network. IEEE Access. https://doi.org/10.1109/ACCESS.2021.3090623Zhou, X., Peng, X., Xie, T., Sun, J., Ji, C., Li, W., & Ding, D. (2021). Fault Anomaly Detection for Microservice Architecture Based on Graph Neural Network. IEEE Access. https://doi.org/10.1109/ACCESS.2021.3090623Bogatinovski, J., Nedelkoski, S., Acker, A., & Kao, O. (2021). Self-Supervised Anomaly Detection from Distributed Traces. IEEE/ACM 21st International Symposium on Cluster, Cloud and Internet Computing (CCGrid). https://doi.org/10.1109/CCGrid51090.2021.00045Bogatinovski, J., Nedelkoski, S., Acker, A., & Kao, O. (2021). Self-Supervised Anomaly Detection from Distributed Traces. IEEE/ACM 21st International Symposium on Cluster, Cloud and Internet Computing (CCGrid). https://doi.org/10.1109/CCGrid51090.2021.00045Chen, H., Chen, P., Wang, B., Yu, X., Chen, X., Ma, D., & Zheng, Z. (2024). Graph neural network based robust anomaly detection at service level in SDN driven microservice system. Computer Networks, 239, 110135. https://doi.org/10.1016/j.comnet.2023.110135Chen, H., Chen, P., Wang, B., Yu, X., Chen, X., Ma, D., & Zheng, Z. (2024). Graph neural network based robust anomaly detection at service level in SDN driven microservice system. Computer Networks, 239, 110135. https://doi.org/10.1016/j.comnet.2023.110135Zhao, H., Wang, Y., Duan, J., Huang, C., Cao, D., Tong, Y., & Zhang, Q. (2020). Multivariate Time-series Anomaly Detection via Graph Attention Network. 2020 IEEE International Conference on Data Mining (ICDM). https://doi.org/10.1109/ICDM50108.2020.00093Zhao, H., Wang, Y., Duan, J., Huang, C., Cao, D., Tong, Y., & Zhang, Q. (2020). Multivariate Time-series Anomaly Detection via Graph Attention Network. 2020 IEEE International Conference on Data Mining (ICDM). https://doi.org/10.1109/ICDM50108.2020.00093Tang, J., Li, J., Gao, Z., & Li, J. (2020). Anomaly Detection in Dynamic Graphs via Transformer. IEEE Transactions on Knowledge and Data Engineering. https://doi.org/10.1109/TKDE.2022.3149801Tang, J., Li, J., Gao, Z., & Li, J. (2020). Anomaly Detection in Dynamic Graphs via Transformer. IEEE Transactions on Knowledge and Data Engineering. https://doi.org/10.1109/TKDE.2022.3149801Deng, A., & Hooi, B. (2021). Graph Neural Network-Based Anomaly Detection in Multivariate Time Series. Proceedings of the AAAI Conference on Artificial Intelligence. https://doi.org/10.1609/aaai.v35i5.16523Deng, A., & Hooi, B. (2021). Graph Neural Network-Based Anomaly Detection in Multivariate Time Series. Proceedings of the AAAI Conference on Artificial Intelligence. https://doi.org/10.1609/aaai.v35i5.16523Le, H.-D., & Park, M. (2024). Enhancing multi-class attack detection in graph neural network through feature rearrangement. Electronics, 13(12), 2404. https://doi.org/10.3390/electronics13122404Le, H.-D., & Park, M. (2024). Enhancing multi-class attack detection in graph neural network through feature rearrangement. Electronics, 13(12), 2404. https://doi.org/10.3390/electronics13122404Boutin, E., et al. (2021). Multi-source Telemetry Fusion for Cloud Incident Detection. IEEE International Conference on Cloud Computing. https://doi.org/10.1109/CLOUD53861.2021.00028Boutin, E., et al. (2021). Multi-source Telemetry Fusion for Cloud Incident Detection. IEEE International Conference on Cloud Computing. https://doi.org/10.1109/CLOUD53861.2021.00028Wu, Y., Dai, H.-N., & Tang, H. (2022). Graph Neural Networks for Anomaly Detection in Industrial Internet of Things. IEEE Internet of Things Journal, 9(12), 9214–9231. https://doi.org/10.1109/JIOT.2021.3094295Wu, Y., Dai, H.-N., & Tang, H. (2022). Graph Neural Networks for Anomaly Detection in Industrial Internet of Things. IEEE Internet of Things Journal, 9(12), 9214–9231. https://doi.org/10.1109/JIOT.2021.3094295Scheinert, D., & Acker, A. (2020). TELESTO: A Graph Neural Network Model for Anomaly Classification in Cloud Services. In Service-Oriented Computing – ICSOC 2020 Workshops, Lecture Notes in Computer Science, vol. 12632, pp. 214–227. Springer. https://doi.org/10.1007/978-3-030-76352-7_23Scheinert, D., & Acker, A. (2020). TELESTO: A Graph Neural Network Model for Anomaly Classification in Cloud Services. In Service-Oriented Computing – ICSOC 2020 Workshops, Lecture Notes in Computer Science, vol. 12632, pp. 214–227. Springer. https://doi.org/10.1007/978-3-030-76352-7_23Mitropoulou, K., Kokkinos, P. C., Soumplis, P., & Varvarigos, E. A. (2024). Anomaly Detection in Cloud Computing using Knowledge Graph Embedding and Machine Learning Mechanisms. Journal of Grid Computing, 22, Article 6. https://doi.org/10.1007/s10723-023-09727-1Mitropoulou, K., Kokkinos, P. C., Soumplis, P., & Varvarigos, E. A. (2024). Anomaly Detection in Cloud Computing using Knowledge Graph Embedding and Machine Learning Mechanisms. Journal of Grid Computing, 22, Article 6. https://doi.org/10.1007/s10723-023-09727-1Marbel, R., Cohen, Y., Dubin, R., Dvir, A., & Hajaj, C. (2024). Cloudy with a Chance of Anomalies: Dynamic Graph Neural Network for Early Detection of Cloud Services’ User Anomalies. arXiv preprint. https://doi.org/10.48550/arXiv.2409.12726Marbel, R., Cohen, Y., Dubin, R., Dvir, A., & Hajaj, C. (2024). Cloudy with a Chance of Anomalies: Dynamic Graph Neural Network for Early Detection of Cloud Services’ User Anomalies. arXiv preprint. https://doi.org/10.48550/arXiv.2409.12726Li, M., et al. (2022). Real-time Anomaly Detection in Cloud Services via Graph Neural Networks. IEEE Internet of Things Journal. https://doi.org/10.1109/JIOT.2022.3168752Li, M., et al. (2022). Real-time Anomaly Detection in Cloud Services via Graph Neural Networks. IEEE Internet of Things Journal. https://doi.org/10.1109/JIOT.2022.3168752Sonani, R., & Govindarajan, V. (2022). A hybrid cloud-integrated autoencoder-GNN architecture for adaptive, high-dimensional anomaly detection in US financial services compliance monitoring. Spectrum of Research.Sonani, R., & Govindarajan, V. (2022). A hybrid cloud-integrated autoencoder-GNN architecture for adaptive, high-dimensional anomaly detection in US financial services compliance monitoring. Spectrum of Research.Li, M., Shu, M., & Lu, T. (2024). Anomaly pattern detection in high-frequency trading using graph neural networks. Journal of Industrial Engineering & Applied Science, 2(6), 77–85. https://doi.org/10.70393/6a69656173.323430Li, M., Shu, M., & Lu, T. (2024). Anomaly pattern detection in high-frequency trading using graph neural networks. Journal of Industrial Engineering & Applied Science, 2(6), 77–85. https://doi.org/10.70393/6a69656173.323430Chaudhary, A., Mittal, H., & Arora, A. (2019). Anomaly Detection using Graph Neural Networks. In 2019 International Conference on Machine Learning, Big Data, Cloud and Parallel Computing (COMITCon), 346–350. IEEE. https://doi.org/10.1109/COMITCon.2019.8862186Chaudhary, A., Mittal, H., & Arora, A. (2019). Anomaly Detection using Graph Neural Networks. In 2019 International Conference on Machine Learning, Big Data, Cloud and Parallel Computing (COMITCon), 346–350. IEEE. https://doi.org/10.1109/COMITCon.2019.8862186Jayaweera, M.P.G.K., Kithulwatta, W.M.C.J.T., & Rathnayaka, R.M.K.T. (2023). Detect anomalies in cloud platforms by using network data: A review. Cluster Computing, 26, 3279–3289. https://doi.org/10.1007/s10586-023-04055-1Jayaweera, M.P.G.K., Kithulwatta, W.M.C.J.T., & Rathnayaka, R.M.K.T. (2023). Detect anomalies in cloud platforms by using network data: A review. Cluster Computing, 26, 3279–3289. https://doi.org/10.1007/s10586-023-04055-1Protogerou, A., Papadopoulos, S., Drosou, A., Tzovaras, D., & Refanidis, I. (2021). A graph neural network method for distributed anomaly detection in IoT. Evolving Systems, 12, 19–36. https://doi.org/10.1007/s12530-020-09347-0Protogerou, A., Papadopoulos, S., Drosou, A., Tzovaras, D., & Refanidis, I. (2021). A graph neural network method for distributed anomaly detection in IoT. Evolving Systems, 12, 19–36. https://doi.org/10.1007/s12530-020-09347-0Jin, M., Koh, H.Y., Wen, Q., Zambon, D., Alippi, C., Webb, G.I., King, I., & Pan, S. (2024). A survey on graph neural networks for time series: Forecasting, classification, imputation, and anomaly detection. IEEE Transactions on Pattern Analysis and Machine Intelligence, 46(12). https://doi.org/10.1109/TPAMI.2024.3443141Jin, M., Koh, H.Y., Wen, Q., Zambon, D., Alippi, C., Webb, G.I., King, I., & Pan, S. (2024). A survey on graph neural networks for time series: Forecasting, classification, imputation, and anomaly detection. IEEE Transactions on Pattern Analysis and Machine Intelligence, 46(12). https://doi.org/10.1109/TPAMI.2024.3443141

The authors declare that there are no conflicts of interest present.