Reaseach On Energy-Aware Real-Time Task Optimization Scheduling Method In Cloud Environment

The characteristics of pay-as-you-go,resources auto-scaling,and the high availability of cloud computing attract more and more users to migrate applications to the cloud.The deployment of many applications on the cloud brings a huge challenge to the cloud scheduler.In cloud computing,task scheduling is one of the key techniques.The effectiveness of task scheduling algorithms is not only related to user experience and quality of service(Qo S),but also has a crucial impact on service providers’ operating costs and load balance among server clusters.Green data centers have been an important research topic in recent years.With the continuous expansion of cloud data centers,high energy consumption has become a stumbling block to the rapid development of cloud computing.In the cloud environment,energy consumption and performance optimization management is an NP-hard multi-objective combinatorial optimization problem.Specifically,the energy-efficient scheduling for realtime tasks is more complicated because such tasks have strict time constraints.Therefore,how to achieve the collaborative optimization of energy consumption and performance needs to be solved urgently.We study the multi-objective combinative optimization scheduling problem of tasks in a cloud computing platform and propose an enhanced ant colony algorithm and deadline-aware backfilling strategy with task completion rate,energy consumption,and load balance as the optimization objectives.The main work is as follows:1、We propose an enhanced ant colony algorithm to achieve a good optimization balance between makespan,energy consumption,and completion rate for the real-time task.Based on the deadline constraints of real-time tasks,we redesign the heuristic information initialization rules and pheromone update rules of the ant colony algorithm so that the algorithm can better converge to a near-optimal solution.2、Then,for real-time task scheduling,a deadline-aware backfilling strategy is proposed to optimize the task scheduling scheme to improve the task completion rate.The backfilling algorithm is performed on the task based on the deadline-sensitive coefficient and improves the task completion rate.3、We propose a two-stage task scheduling framework based on the enhanced ant colony algorithm and deadline-aware backfilling strategy to solve and optimize the task scheduling scheme by considering the makespan,energy consumption,and task completion rate.Finally,we conduct extensive experiments to validate the effectiveness of our scheduling algorithm based on Cloud Sim,a common open-source cloud simulation platform in the area of cloud computing.