Exploration on Resource Scheduling Optimization Strategies in Cloud Computing Environment

Abstract

The rapid expansion of cloud computing has brought opportunities and difficulties for resource schedule optimization to enhance productivity and decrease expenses. This study offers a thorough analysis of resources planning optimisation techniques in cloud computing settings, emphasizing the most recent developments and approaches. First, we go over the basic ideas of resource scheduling, such as resource sharing, load balancing, and task distribution. The task scheduler's execution-time mappings among the future demand and cloud resources makes application planning one of the major difficulties in cloud computing. By decreasing makespan and improving resource utilization, an effective scheduling system is required to schedule the diversified workload and enhance performance metrics. Numerous scheduling methods that were previously in use only considered makespan and utilization of resources metrics, ignoring other important factors that have an immediate effect on cloud service performance, such as energy consumption and migration time. To address the problems, the authors have developed the Hybridize Whale Optimization Algorithm (H-WOA), a nature-inspired multi-objective task scheduling algorithm that can make scheduling decisions at runtime depending on the availability of cloud resources and impending workload requirements. Furthermore, the suggested method distributes the resources according to task priorities and end users' budgets. The workload for the proposed H-WOA technique, which is based on the Cloudsim toolkit, is created by creating datasets (da01, da02, da03, da04) with various task densities and workload records from NASA (da05, da06) and HPC2N (da01, da02) parallel workload repositories. An extended experiment's findings demonstrate that the suggested H-WOA technique enhanced the important variables and performed better than alternative baseline policies.