Optimizing Cloud Data Centers for Environment Sustainability through Performance-Aware Virtual Machine Allocation

Cloud computing faces challenging issues concerning high-energy consumption and service-level agreement (SLA) violations, particularly as the demand for computational power continues to grow, necessitating large-scale cloud data centers. Existing energy-aware approaches tend to focus on performance optimization during VM migration phases but it can be effectively maintained at VM allocation level as preprocessing. Thereby, this paper proposed Energy Efficient SLA-aware virtual machine Allocation approach using a metaheuristic Firefly algorithm (FFO-EESA). The experimental findings showcase promising outcomes in the context of dynamic workloads and varying numbers of VMs (ranging from 1 to 290) allocated on individual hosts. The proposed method’s experimental results were obtained through computer-based simulations conducted in MATLAB version 2017(a). The outcomes showed that, there was a reduction of 12.34% in migrations and a saving of 4.36% in idle hosts with the FFO-EESA approach. proposed algorithm resulted in an average energy savings of 4.39% compared to the FF and MBFD techniques.