A Load Balancing Algorithm Designed for Data Centres to Enhance the Efficiency of Cloud Computing Tasks

Despite the extensive research conducted in the field of Cloud Computing, challenges still persist regarding workload balancing in cloud-based applications, particularly within the Infrastructure as a Service (IaaS) model. Efficient task allocation is vital in cloud computing due to the limited availability of resources and virtual machines. IaaS is a model that manages the backend infrastructure, including servers, data centres, and virtual machines. Cloud Service Providers must ensure optimal service delivery performance in these models to prevent issues like host overload or underload, which can lead to increased execution times or machine failures. Task Scheduling plays a significant role in load balancing, and it is essential that task scheduling aligns with the requirements outlined in the Service Level Agreement (SLA), a document provided by cloud developers to users. Key SLA parameters, such as deadlines, are considered in the load balancing algorithm. The proposed algorithm aims to optimize resource usage and enhance load balancing by taking into account Quality of Service (QoS) task parameters, VM priorities, and resource allocation. This load balancing algorithm addresses the identified challenges and fills the current research gap based on findings from the literature. Results indicate that the proposed algorithm achieves an average resource utilization of 90% compared to the existing Dynamic LBA algorithm, while also demonstrating improved performance in terms of reduced execution time and makespan.