An efficient load balancing in cloud computing using hybrid Harris hawks optimization and cuckoo search algorithm

Cloud computing has rapidly emerged as a burgeoning research field in recent times. However, despite this growth, a comprehensive examination of this domain reveals persistent issues in the application of cloud-based systems concerning workload distribution. The abundance of resources and virtual machines (VMs) within cloud computing underscores the importance of efficient task allocation as a critical process. Within the infrastructure as a service (IaaS) architecture, load balancing (LB) remains a pivotal but challenging task. The occurrence of overloaded or underloaded hosts/servers during cloud access is undesirable, as it leads to operational delays and system performance degradation. To address LB issues effectively, it is imperative to deploy a proficient access scheduling algorithm capable of distributing tasks across the available resources. A novel approach was introduced by combining the Harris hawk's optimization and cuckoo search algorithm (HHO-CSA), with a specific focus on critical service level agreement (SLA) parameters, particularly deadlines, to uphold LB in a cloud environment. The primary objective of the hybrid HHO-CSA methodology is to provide task attributes, resource allocation, VMs prioritization, and quality of service (QoS) to clients within cloud computing applications. The outcome analysis reveals that the proposed hybrid HHO-CSA algorithm results in a resource utilization reduction of 52%, with an execution time of 529.84 ms and a makespan of 638.88 ms. These values outperform those of existing SLA-based LB algorithms. Effective task scheduling plays a pivotal role in ensuring the seamless execution of tasks within a cloud system, while LB significantly aligns with the SLAs available to users. Drawing insights from the existing literature, the suggested hybrid HHO-CSA method addresses the research gap by effectively mitigating the challenges.