Forecasting Energy Efficiency in the Public Sector through Cloud Data Centers Machine Learning Consolidation

Cloud computing provides on-demand access to distributed resources, services, and applications across the network. One of the key challenges in modern CDCs is efficient energy management due to their large scale of infrastructure along with huge power consumption. The rapid expansion of cloud technology has a deep impact upon energy usage and environmental sustainability due to increasing resource demands. Energy-efficient strategies are explored as solutions to this issue that utilize machine learning techniques. ML techniques can be used to optimize energy consumption in CDCs by analyzing data patterns, predicting CPU utilization, identifying system overload or underload, and managing VM allocation and migration efficiently. This paper presents various ML-driven approaches for minimizing energy consumption with preserved SLA. The results show that the ML-based optimization techniques save 1.6% to 88.5% of energy as compared to the traditional benchmarks based on the system parameters and configurations.