| With the development of technology,Cloud Computing has penetrated into all aspects of life as a new computing model.It has convenient and efficient service.And it benefits from Cloud Data Center.It is the infrastructure of Cloud Computing and it provides the basic support for Cloud Computing Services.Reasonable resource management can improve the efficiency of Cloud Computing Services in Cloud Data Center.Therefore,the resource management plays a crucial role in the research of Cloud Data Center which draw a considerable attention from scholars both academic and industry.Currently,research scholars have focused on workflow scheduling in Cloud Data Center.Because of the increase in virtualized resources and user requests online.In the process of resource management,Cloud Data Center have some problems such as unbalanced load,low task processing efficiency,and high cost.Therefore,in this paper,it analyzes and models the low efficiency of workflow scheduling and the joint scheduling problem of cost optimization in Cloud Data Center.This paper proposes an efficient particle swarm optimization scheduling algorithm and it is based on adaptive weight.It also designs a workflow joint scheduling algorithm and it is based on cost optimization.The experimental results prove the effectiveness of the proposed algorithm.The main research work of this paper is as follows:(1)Aiming at the low efficiency of workflow scheduling in Cloud Computing,this paper proposes a particle swarm optimization workflow scheduling algorithm based on adaptive weights.Firstly,this paper describes the problem that the task scheduling problem in cloud resources.It defines the relevant parameters in this paper.At the same time,this paper constructs an objective function that takes the total completion time of the task as the optimization object.Secondly,this paper optimizes the inertia weight factor and the fitness function of the particles according to the number of iterations.Finally,the experimental results prove that the proposed algorithm is efficient in this paper.Because when it runs on small-scale and large-scale task sets,and it can reduce the task completion time and improve the utilization of virtual machine resources.(2)Aiming at the joint scheduling problem of excessive overhead,this paper designs a twostage workflow joint scheduling algorithm and it is based on ant colony system and greedy thinking.Firstly,According to the characteristics of the workflow,this paper constructs a workflow model,According to multiple types of virtual machines,we also consider that merging virtual machines of the same type,which can avoid the problem of excessive overhead of virtual machine deployment.We build model of virtual and physical machines.Secondly,in the first stage,we address the constructed model that this paper is based on ant colony system algorithm.It designs heuristic functions and pheromone update rules to minimize the completion time of the workflow.And then the workflow is mapped to the virtual machine.In the second stage,under the premise of ensuring the completion of the workflow,the main feature of this method is that it takes into account multiple types of virtual machines,so as to avoid the problem of high cost of virtual machine deployment.This paper designs a virtual machine-to-physical machine deployment method and it is based on greedy thinking.The experimental results prove that the proposed this method improves the time efficiency of joint scheduling.And comparing with the traditional algorithms,it reduce joint scheduling time and saving cost. |
