为了解决在云环境下,由于作业规模的不断增大和种类的不断增多而导致的调度执行混乱的问题,提出了利用决策树C4.5算法对调度作业进行分类的方法。该方法采用自顶向下的递归方式,将一组无序的数据整理成类似于流程图的结构,并对所得数据进行最大“差异”切分,最终得到大规模问题的简单集。将该方法应用于云调度算法中,参数反演结果反映了在执行效率和客户满意度等方面,较传统的调度算法都有较大的提高,从而也验证了该方法的有效性。
As a result of the increasing sizes and types of the jobs in the cloud environment, the job scheduling execution becomes more and more confusion. To solve the chaos, a method based on the C4.5 algorithm of decision tree to classify the jobs before scheduling was intro duced,which uses the top-down reeursive approach to organize a set of unordered data into flow chartqike structure,and to segement the data based on the atmost "difference" to get the simple set of the large-scale problems. The method was used in the cloud scheduling algorithm and the parameters showed a higher executive efficiency and a better customer satisfaction compared with the traditional method. Thus, the results illustrated the validity of this method.