Leveraging Reinforcement Learning for Autonomous Cloud Management and Self-Healing Systems

Nisher Ahmed (College of Technology & Engineering, Westcliff University, Irvine, California, USA.)
Md Emran Hossain (College of Technology & Engineering, Westcliff University, Irvine, California, USA.)
S M Shadul Islam Rishad (College of Technology & Engineering, Westcliff University, Irvine, California, USA.)
Arafath Bin Mohiuddin (College of Technology & Engineering, Westcliff University, Irvine, California, USA.)
Md Imran Sarkar (College of Technology & Engineering, Westcliff University, Irvine, California, USA.)
Zakir Hossain (College of Engineering and Computer Science, California State University, Northridge,California, USA.)

Jurihum
Vol. 1 No. 6 (2024): JURIHUM : Jurnal Riset dan Humaniora

This work investigates and enhances innovative methods for autonomous cloud management with artificial intelligence, specifically self-healing and self-optimization. The study uses AI based anomaly detection, predictive maintenance and automated recovery to derive self-healing. To self-optimize, it employs machine learning algorithms to analyze existing workload patterns, anticipate resource utilization demand, and adjust resources dynamically. These techniques are tested and validated in a simulated cloud environment in terms of performance metrics like response time, throughput, and resource utilization.