A Metaheuristic-Based Scientific Workflow Scheduling in IaaS Cloud

As cloud data centres expand and provide more services, they consume more energy and cause challenges for the environment. To address this, there is a focus on energy-saving scheduling approaches in cloud computing. This study proposes a novel method for scheduling workflows in cloud computing that is based on the MaxUtil framework. The proposed method integrates the well-known flower pollination algorithm (FPA), a meta-heuristic algorithm derived from natural phenomena. The suggested scheduling approach aims to reduce energy usage and workflow processing time. The suggested method consists of two stages: (i) job assignment to virtual machines (VMs) that are available, and (ii) task scheduling based on optimal criteria. Extensive research has been undertaken throughout five scientific procedures from diverse domains. Tests on various scientific workflows show that this method performs better than other common scheduling methods like PSO, GSA, and GA in most cases.