Research On Cloud Workflow Scheduling Method Based On Multi-QoS Objective

With the rapid development of cloud computing and the application of workflow technology in cloud computing,the issue of cloud workflow task scheduling has attracted widespread attention.How to satisfy the Quality of Service(Qo S)such as execution time,system utilization,and cost is a problem that needs to be solved in the cloud workflow task scheduling problem.In recent years,a large number of methods have been proposed for each optimization goal in the field of cloud workflow scheduling.Some of these methods only consider a single target to make its goal of scheduling optimal,but do not consider the multi-Qo S problem of cloud workflow scheduling.In view of the problem that the existing methods fail to consider the goal of quality of service,this paper considers the attributes of cloud workflow tasks and resource nodes into the scheduling model,and then introduces novelty ranking and similarity measure to solve the problem.Finally,a series of experiments are conducted to confirm the efficiency and rationality of our proposal.In detail,the innovation and main work of this paper can be summarized as follows.Firstly,in order to solve the problem of high efficiency and user cost in cloud environment,in this paper,a cloud services workflow scheduling algorithm based on novelty ranking strategy and multi-Qo S objective(RMO-CWS)is proposed.And this algorithm initially takes into consideration the frequency of resource nodes performing on the task,the latency time of task and the execution time as factors to put into the recommended model trained by a simulated annealing algorithm with the priority factor calculated.And then,recommend tasks to resource nodes that it has not previously distributed to,that means,high novelty.Finally,the scheduler works according to the priority scheduling factor table and updates them.The results show that the proposed method reduce the execution time of tasks,and make aggregate indicator on task execution time and system utilization better.Secondly,aiming at the problem of system efficiency and user cost in similar task scheduling in cloud environment,in this paper,a cloud workflow scheduling algorithm based on similarity measure and multi-Qo S objective(SM-CWS)is proposed.This algorithm introduces the similarity measure knowledge.Then the workflow scheduling problem is re-modeled,and the attributes of task and virtual machine are taken as recommended attributes to be considered.And similar tasks are scheduled to be executed on the recommended virtual machines to reduce the resource scheduling time,so as to optimize the task scheduling.Make multi-objective(task execution time,system performance,cost)cloud workflow scheduling problems are optimized.Thirdly,through multi-group experimental analysis with open source simulation software Cloud Sim and workflow models,the proposed method is verified to be better than other algorithms in task total execution time,task average execution time and system utilization.In addition,the task execution time of SM-CWS is about 18.083% higher than that of Max-Min algorithm and about 20.834% higher than that of Min-Min algorithm.The system utilization of SM-CWS is about 19.962% higher than that of the Max-Min algorithm and about 9.336% higher than that of the Min-Min algorithm.