Authors: Farhana Yasmin
Abstract: The growing complexity of hybrid infrastructures, combining on-premises and cloud systems, demands advanced monitoring frameworks capable of handling dynamic, large-scale environments. Traditional rule-based monitoring solutions often fail to detect subtle or novel anomalies that emerge in such heterogeneous ecosystems. Artificial Intelligence (AI)-enhanced system monitoring has revolutionized anomaly detection by integrating machine learning, predictive analytics, and automation into network and system surveillance. This review explores the mechanisms, benefits, and challenges of AI-driven anomaly detection in hybrid infrastructures. It discusses how AI techniques such as deep learning, unsupervised clustering, and neural networks improve accuracy, speed, and contextual understanding in detecting irregular patterns. Furthermore, the paper evaluates hybrid monitoring architectures, data-driven models, and predictive capabilities that support proactive maintenance and security resilience. The review concludes by emphasizing AI's transformative role in achieving intelligent, adaptive, and self-healing IT operations within hybrid environments.
DOI: http://doi.org/10.5281/zenodo.17840111
