Machine Learning-Based Anomaly Prediction for Proactive Monitoring in Data Centers: A Case Study on INFN-CNAF DOI Creative Commons
Andrea Asperti,

Gabriele Raciti,

Elisabetta Ronchieri

et al.

Applied Sciences, Journal Year: 2025, Volume and Issue: 15(2), P. 655 - 655

Published: Jan. 11, 2025

Anomaly prediction in time series is crucial for ensuring the stability and security of data centers, especially scientific contexts such as INFN-CNAF, National Center Research Development Information Communication Technology Institute Nuclear Physics. At large volumes heterogeneous critical to international experiments are managed using dedicated monitoring systems. To ensure continuous availability, artificial intelligence solutions being explored detect anomalies predict potential failures proactively. This work presents a machine learning-based approach automatic anomaly operational metrics INFN-CNAF’s WebDav service. We evaluate several methods, including Long Short-Term Memory, Random Forest, various neural networks, assessing their Accuracy sensitivity distinguishing normal from anomalous behaviors. The results demonstrate effectiveness these not only predicting but also pinpointing areas within monitored metrics. contributes more proactive IT resource enhances center management efficiency.

Language: Английский

Transportation mode detection through spatial attention-based transductive long short-term memory and off-policy feature selection DOI Creative Commons

Mahsa Merikhipour,

Shayan Khanmohammadidoustani,

Muhammad Daud Abbasi

et al.

Expert Systems with Applications, Journal Year: 2024, Volume and Issue: 267, P. 126196 - 126196

Published: Dec. 25, 2024

Language: Английский

Citations

5
Machine Learning-Based Anomaly Prediction for Proactive Monitoring in Data Centers: A Case Study on INFN-CNAF DOI Creative Commons
Andrea Asperti,

Gabriele Raciti,

Elisabetta Ronchieri

et al.

Applied Sciences, Journal Year: 2025, Volume and Issue: 15(2), P. 655 - 655

Published: Jan. 11, 2025

Anomaly prediction in time series is crucial for ensuring the stability and security of data centers, especially scientific contexts such as INFN-CNAF, National Center Research Development Information Communication Technology Institute Nuclear Physics. At large volumes heterogeneous critical to international experiments are managed using dedicated monitoring systems. To ensure continuous availability, artificial intelligence solutions being explored detect anomalies predict potential failures proactively. This work presents a machine learning-based approach automatic anomaly operational metrics INFN-CNAF’s WebDav service. We evaluate several methods, including Long Short-Term Memory, Random Forest, various neural networks, assessing their Accuracy sensitivity distinguishing normal from anomalous behaviors. The results demonstrate effectiveness these not only predicting but also pinpointing areas within monitored metrics. contributes more proactive IT resource enhances center management efficiency.

Language: Английский

Citations

0