Abstract The efficiency of cloud resource allocation and load balancing is a critical challenge with the increasing dynamic workloads. This leads to performance degradation and load imbalance. This paper proposes an enhanced dynamic grey wolf optimization (EDGWO) algorithm that integrates the application of a dynamic elite archive blending to improve solution diversity and convergence, and a stochastic Levy flight perturbation strategy that aids the algorithm to avoid stagnation in local optima. The EDGWO was rigorously evaluated on the CEC2020 benchmark suite (F1–F10), achieving the best performance in 6 out of the 10 test functions, and second best in the remaining 4 across best fitness, mean fitness and standard deviation. A CloudSim simulation assessing VM load balance and utilization demonstrated EDGWO’s better performance over GWO, PSO, GA, and ABC, achieving an average of 25% improvement in terms of resource utilization and load balancing. The findings highlight the effectiveness and robustness of EDGWO for a performance driven cloud resource management.
Aburinya et al. (Tue,) studied this question.