Hybrid Genetic Algorithm for Efficient Load Balancing Through Virtual Machine Migration in Cloud Computing Environments

Cloud computing environments face significant challenges in resource management and load balancing across distributed virtual machine (VM) hosts. This paper presents a Hybrid Genetic Algorithm (HGA) approach for optimizing VM migration and load balancing in cloud computing environments. The proposed HGA integrates infeasible solution revamping and local optimization techniques to enhance the standard Genetic Algorithm (GA). Experimental results demonstrate that HGA achieves 98.2% accuracy compared to 87.7% for the MultiObjective Artificial Bee Colony with Q-learning (MOABCQ) algorithm, representing a 10.5% improvement. Additionally, HGA reduces energy consumption by 15.1% (from 90.6 J to 78.7 J) and migration cost by 12.9% (from 93297 to 82656 units). The proposed approach ensures efficient resource utilization, improved Quality of Service (QoS), and enhanced system stability in cloud computing environments.

Keywords: Load Balancing, Virtual Machine Migration, Genetic Algorithm, Cloud Computing, Optimization, Resource Management.