An improved Firefly Algorithm based Task Scheduling in Cloud Computing for Effective Resource Utilization

Computing possessions may be dynamically scaled up and down at little cost in the cloud. In cloud computing, well-organized job scheduling is regarded the most important difficulties. For large tasks, task scheduling is a difficult problem since it is an NP-complete problem. If you're using a cloud computing platform, you may have to schedule many jobs on multiple virtual machines (VMs) to minimize the time it takes to complete each activity, as well as maximize the amount of resources utilized. This research proposes a multi-objective Firefly cyclic randomization (FCR) based job allocation method for dealing with the large number of cloud user requests. Randomization and the modified firefly algorithm are used to improve the task scheduling process of the suggested approach. FCR takes into account many objectives and discoveries the best technique of job allocation based on DI, time to completion and throughput. In order to assess FCR's performance, we compared its throughput, average response time, DI, and makespan metrics to those of many existing methods. The experiments' findings demonstrate that FCR outperforms other task allocation algorithms.