Energy-Efficient Virtual Machine Placement in Cloud Data Centers Using Reinforcement Learning-Enhanced Adaptive Greedy Dingo Optimization Algorithm

This paper presents a hybrid intelligent framework combining Adaptive Greedy Dingo Optimization Algorithm (AGDOA) with Reinforcement Learning (RL) for energy-efficient Virtual Machine (VM) placement in cloud data centers. The proposed approach integrates metaheuristic optimization, machine learning-based workload prediction, and adaptive security mechanisms to address the NP-hard VM placement problem. Experimental evaluation on diverse cloud configurations demonstrates 33% improvement in overall system efficiency, 8.5× reduction in computation time, and 20.66% enhancement in confidential data handling rates compared to state-of-the-art methods (EFHE-SV, PARM, SCP). The framework successfully handles 24 servers with 20 VMs while maintaining 98.5% resource utilization efficiency and reducing power consumption by 1.42E+05W. Statistical significance testing confirms p-value< 0.001 across all performance metrics. The proposed RL-enhanced AGDOA demonstrates superior convergence properties and robust scalability across varying cloud infrastructure configurations.

Keywords: Virtual Machine Placement, Cloud Computing, Reinforcement Learning, Metaheuristic Optimization, Energy Efficiency, Adaptive Algorithms, Resource Allocation, Cloud Security