Study On The Scheduling Algorithm For Workflow Task In A Cloud Computing System

The scheduling algorithm has provided an effective solution for cloud service in a cloud computing system.However,the existing scheduling algorithms determine the scheduling plan of the workflow task prior to actual scheduling.They ignore the workflow task scheduling result affected by the actul scheduling environment state change(including the change of the resource server state,the dynamic change of network bandwidth,etc.)and the dynamic change of the workflow path in the actual scheduling process.Aiming to the above-mentioned defects of the cloud computing scheduling algorithm,the scheduling algorithm for workflow task in a cloud computing System is proposed.The research work is as follow:To solve the problem of task scheduling in the cloud computing system,this paper proposes a scheduling algorithm for cloud computing based on the driver of dynamic essential path(DDEP).This algorithm applies a predecessor-task layer priority strategy to solve the problem of constraint relations among workflow tasks.The strategy assigns different priority values to every workflow task based on the scheduling order of workflow task as affected by the constraint relations among workflow tasks,and the workflow task list is generated by the different priority value.To address the scheduling order problem in which workflow tasks have the same priority value,the dynamic essential long path strategy is proposed.This strategy computes the dynamic essential path of the pre-scheduling workflow tasks based on the actual computation cost and communication cost of workflow task in the scheduling process.The workflow task that has the longest dynamic essential path is scheduled first as the completion time of task graph is indirectly influenced by the finishing time of workflow tasks in the longest dynamic essential path.Finally,we demonstrate the proposed algorithm via simulation experiments using Matlab tools.The experimental results indicate that the proposed algorithm can effectively reduce the task Makespan in most cases and meet a high quality performance objective.This paper proposes a scheduling algorithm to solve the problem of task scheduling in a cloud computing system based on a weighted time-series network bandwidth model.This algorithm converts the scheduling problem with communication changes into a directed acyclic graph(DAG)scheduling problem for existing fuzzy communication workflow tasks,i.e.,the scheduling problem for a communication-change DAG(CC-DAG).The CC-DAG contains both computation workflow tasks and communication workflow tasks.First,this paper proposes a weighted time-series network bandwidth model to solve the indefinite processing time(cost)problem for a fuzzy communication workflow task.This model can accurately predict the processing time of a fuzzy communication workflow task.Second,to address the scheduling order problem for the computation workflow tasks,a dynamic pre-scheduling search strategy(DPSS)is proposed.This strategy computes the essential paths for the pre-scheduling of the computation workflow tasks based on the actual computation costs(times)of the computation workflow tasks and the predicted processing costs(times)of the fuzzy communication workflow tasks during the scheduling process.The computation workflow task with the longest essential path is scheduled first because its completion time directly influences the completion time of the task graph.Finally,we demonstrate the proposed algorithm via simulation experiments.The experimental results indicate that the proposed algorithm can effectively reduce the makespan in most cases and can achieve high performance.Aiming to the problem of the task scheduling result affected by the state change of server in the cloud computing system,the scheduling algorithm for the cloud computing system based on the dynamic feedback of server state(DFSS)is proposed.The algorithm adopts the proposed sliding window model to reflect the server busy state in real time.To accurately predict the state of server,the algorithm uses the proposed weight curve model of time window and the average utilization rate model to compute the average utilization rate in the period of time of the future.The average utilization rate denotes the busy state of server at the different time periods.The dynamic rank value of every workflow task could be got by the proposed server feedback weight strategy based on the average utilization rate.All workflow tasks are sorted by their dynamic rank value in descending order,then adopts the earliest finish strategy to determine the start processing time and server according to the order.Finally,we demonstrate the proposed algorithm via simulation experiments by Matlab tool.The experiment result indicates the proposed algorithm can not only effectively reduce the task Makespan in most cases,but also have high quality performance objective.To solve the problem of the deadline-constrained task scheduling in the cloud computing system,this paper proposes a deadline-constrained scheduling algorithm for cloud computing based on the driver of dynamic essential path(Deadline-DDEP).According to the change of the dynamic essential path of each workflow task in the scheduling process,the dynamic sub-deadline strategy is proposed.The strategy assigns different sub-deadline values to every workflow task to meet the constraint relations among workflow tasks and the user's defined deadline.The strategy fully considers the dynamic sub-deadline affected by the dynamic essential path of workflow task in the scheduling process.To address the problem of selecting server for each workflow task,the quality assessment of optimization cost strategy is proposed,on basis of the sub-deadline urgency and the relative execution cost in the scheduling process,the strategy selects the server that not only meet the sub-deadline but also obtains more lower execution cost.In this way,the algorithm will make the task graph complete within its deadline,and minimize the total execution cost.Finally,we demonstrate the proposed algorithm via simulation experiments using Matlab tools.The experimental results show that,the proposed Deadline-DDEP produced remarkable performance improvement rate on the total execution cost that ranges between 10.3% to 30.8% while meeting the deadline constraint.In view of the experimental results,the proposed algorithm provides better-quality scheduling solution that is suitable for scientific application task execution in the cloud computing environment than IC-PCP,DCCP and CD-PCP.