A Threshold-Based Hybrid GWO–PSO Framework with Trust-Aware Security for Efficient Load Balancing in Cloud Computing

Abstract

Cloud computing is becoming increasingly complex and is constantly evolving as it uses a variety of resources to perform tasks. The challenge lies in monitoring cloud providers and ensuring they use the appropriate amount of computer resources for each task. This study presents an intelligent workload allocation model created using a hybrid optimisation approach that utilises Grey Wolf Optimisation (GWO) and Particle Swarm Optimisation (PSO). This solution uses a nonlinear composite model to track the resources allocated to a virtual machine (VM) and adjust the thresholds that determine the VM's system state, ensuring processing continues smoothly. To speed up the process of assigning jobs and make the most of the computer resources used for this purpose, a priority-based approach will be implemented. This will enable work to be shared between computers, ensuring that the servers delivering services do not become overloaded. A trust-based virtual machine filtering provides a lightweight security-awareness layer that ensures dependable task allocation and efficient load balancing.Our experimental results show that the proposed workload distribution method surpasses existing load balancing techniques for each participating cloud service provider across all quality metrics (e.g. total make span, energy consumption, throughput and degree of distribution imbalance). This study shows that the GWO-PSO hybrid optimisation framework can provide fast, flexible load balancing solutions for cloud computing environments.