Research On Task Scheduling Algorithm In Time-oriented Grid Environment

Cloud computing,which has evolved from grid computing,has become a hot technology and research hotspot in recent years due to its powerful computing power and flexible resource management.As a basic technology,grid technology has been continuously researched and developed.The task scheduling algorithm as the core content determines the overall scheduling performance of the grid and the processing efficiency of the system.Therefore,relevant research on task scheduling algorithms is necessary.This thesis analyses the importance and significance of task scheduling algorithm at first,then from the model and process of task scheduling in grid environment makes a comparative analysis of the static and dynamic algorithms.At the same time,this article analyzed the core idea of MCT algorithm and Min-Min algorithm,and the characteristics and shortcomings of the two algorithms;also make compare from the advantages and disadvantages of the two algorithms and make a final summary.Analysis and summary: MCT algorithm in the scheduling process to find the local optimal solution,lack of overall consideration;Min-Min scheduling process will be biased towards small task processing,both algorithms have a certain space for improvement.By studying the related content and aiming at the static scheduling algorithm in grid environment,this paper proposes to shorten the completion time by analyzing and adjusting the task group.Under the condition of restricting the length of each task group,starting with the different proportion of long task and short task,combining with the calculation of weight factors,the task sequence is adjusted compulsively and then the scheduling is calculated.In the experiment,the differences between the results adjusted by weight and the original results under different tasks were compared,and the analysis and summary were made.After obtaining the influence of weight factors on the scheduling results,the optimal selection mechanism and scheduling mechanism combined with task analysis are proposed.The analysis of the task array and the calculation of related parameters are used.The conclusion is obtained,the most suitable adjustment method is selected and the appropriate scheduling process is selected by using parameters,so as to optimize the scheduling result with the completion time as the set target.The innovation of this paper lies in the internal analysis of the task group,combined with the length of the task,the proportion of the length of the task and the value of the weight of the factors,to study the impact on the task completion time;on this basis,the optimal weight selection mechanism and scheduling mechanism are implemented.Combined with the analysis of the task array and the calculation of multi-parameters,the optimal selection is made under different task conditions to optimize the scheduling results.