Scalable and Optimized Load Balancing in Cloud Systems: Intelligent Nature-Inspired Evolutionary Approach

The optimal system performance depends on efficient scheduling of numerous virtualized resources which Cloud computing orchestrates. Organizations using cloud computing require efficient task scheduling to achieve optimal system performance because the platform includes multiple virtualized resources. This paper proposes a novel Hybrid Lyrebird Falcon Optimization Algorithm (HLFOA) for global exploration and the Falcon Optimization Algorithm (FOA) for local exploitation. Through HLFOA virtual machine (VM) tasks become better distributed across sites while achieving minimum makespan together with reduced power usage and enhanced CPU resource utilization. Performance analysis with CloudSim 4.0 simulation proves that HLFOA is more efficient than baseline methods as PSO. At 100 tasks, HLFOA achieves a makespan of 299 units, compared to PSO's 513 units, and at 500 tasks, it reduces makespan to 2015 units, while PSO reaches 3868 units. The adoption of HLFOA improves both system energy consumption efficiency and processor utilization levels. HLFOA shows promise as a scalable and effective solution for cloud load balancing, which enables robust optimization of cloud resource allocation.