Mantaray modified multi-objective Harris hawk optimization algorithm expedites optimal load balancing in cloud computing

Abstract

Task scheduling in the cloud is a difficult optimization challenge. The cloud system is assigned with a specific load based on the cloud architecture and user demands. However, both underloaded and overloaded situations result in a variety of system failures in terms of power consumption, machine failure, and so on. As a result, task-load balancing on virtual machines (VMs) is considered as an important part of cloud task scheduling. The present study proposes a dynamic load balancing algorithm based on the hybrid optimization algorithms named as Mantaray modified multi-objective Harris hawk optimization (MMHHO). The hybridization process updates the search space of Harris Hawk Optimization (HHO) by utilizing the Manta Ray Forging Optimization (MRFO) algorithm by considering the cost, response time, and resource utilization etc. The hybrid scheme, proposed in the present study, improves the system performance by enhancing the VMs throughput, balancing the load between the VMs, and sustaining the balance among priorities of tasks by adjusting the waiting time of the involved tasks. The proposed MMHHO based load balancing algorithm is implemented in CloudSim tool. The effectiveness of the suggested algorithm has been analyzed in terms of various parameters and compared with other existing algorithms. The simulation results show that the suggested MMHHO load balancing scheme outperforms other algorithms.