Research On Task Scheduling In Cloud Computing Data Center

In today’s society, cloud computing has become more and more popular. In many large enterprises or departments and units, have established their own cloud data center. Task scheduling in cloud computing has become the focus of academic research. How to further optimize the scheduling strategy and scheduling algorithm to meet the needs of different target becomes more and more important.Based on the full reading a lot of literature, for different types of task of modeling and deep analysis, and put forward the independent task scheduling algorithm based on energy consumption and the dependent task scheduling algorithm based on the completion time.Scheduling algorithm in the independent task, using the task merging subsets of thought to become the task scheduling, and the computational task nodes subset assigned to optimal performance on executive. In the task subset, by removing the time interval between the task to reach the purpose of energy saving.In the research task dependency, through carries on the analysis to the overall factors influencing the task execution, design and calculation method of the new priority. In the distribution phase of the task, for all the parent task execution of tasks into consideration, through the task redundancy replication thoughts, to reduce the communication cost of the task set. Through the study found that, to a certain extent, can be set for the task of task execution time in advance, so as to shorten the time of completion of the task set.Finally, through the CloudSim simulation tool, the above algorithm is verified. Verification results show that, the algorithm in comparison with other algorithms under can effectively save energy and shorten the time of completion, so as to further enhance the task scheduling and execution efficiency of computing nodes.