<![CDATA[Objective: In this research we explore the use of AI-driven monitoring and anomaly detection for analyzing risks in AWS and Azure environments, with an exploratory study on the Cloud Data Center Workload Dataset. The research will explore the following objectives: understanding the behavior of cloud workloads, detecting abnormal workload behavior, investigating the correlation between cloud monitoring parameters, and evaluating the risks of the operational environment of cloud infrastructure. Method: The methodology is based on an exploratory research approach with data preprocessing, exploratory data analysis, statistical analysis, and implementation of the monitoring framework using AI and R Studio. To interpret the behavior of the cloud, multiple visualizations were developed: workload distribution, resource utilization analysis, anomaly monitoring dashboard, correlation network, and operational risk dashboard. Results: The results indicate that changes in CPU usage, network traffic, memory consumption, processor temperature, or virtual machine activity affect the performance of the cloud and uncover potential points of vulnerability for operation. Moreover, with AI-powered monitoring one can easily detect abnormal patterns of resource consumption and continuously assess operational risk by combining the monitoring parameters into a composite risk index. In conclusion, the study demonstrates that AI support in monitoring bolsters cybersecurity management practices, enhances infrastructure visibility, aids in early detection of anomalies, and aids in proactive management of operational risks across AWS and Azure cloud platforms. Novelty: This study integrates AI-driven monitoring and anomaly detection with cloud workload analysis to evaluate operational risks across both AWS and Azure environments using the Cloud Data Center Workload Dataset.]]>
Copyrights © 2024
