ENERGY-EffICIENT JOB SCHEDULING IN GREEN CLOUD COMPUTING USING NEURAL NETWORKS

Cloud computing has revolutionized modern digital infrastructure but comes with a significant energy cost due to large-scale data center operations. This paper presents a hybrid scheduling framework that integrates Artificial Neural Networks (ANNs) with the Multi-Objective Evolutionary Algorithm based on Decomposition (MOEA/D) to achieve energy-efficient job scheduling in green cloud environments. The ANN predicts optimal resource mappings based on real-time and historical workload metrics, while MOEA/D refines these mappings by balancing energy consumption and performance metrics such as response time. Simulation results using CloudSim demonstrate up to 15% reduction in energy usage, improved server consolidation, and lower SLA violation rates compared to traditional algorithms like FCFS and Round-Robin. The framework demonstrates strong scalability and adaptability, providing a robust solution for sustainable cloud computing infrastructures.