改进的DBSCAN聚类算法在云任务调度中的应用

Application of Improved DBSCAN Clustering Algorithm in Task Scheduling of Cloud Computing

针对云计算环境中任务调度中存在的执行效率低的问题,提出了一种基于改进的基于密度的聚类算法(DBSCAN)的云任务调度策略.首先使用改进的基于密度的聚类算法DBSCAN对云任务进行聚类,然后与已经分类的资源进行匹配,解决资源与任务匹配程度低的问题.实验结果表明,对任务进行聚类后进行任务调度,任务在终端上的平均执行时间减少了大约35.2%,任务的调度时间也有了明显减少.

Cloud scheduling strategy based on improved density-based spatial clustering of applications with noise( DBSCAN) clustering algorithm was proposed to solve the problem of low efficiency of task scheduling in the implementation of cloud computing environment. Firstly,an improved DBSCAN clustering algorithm was used to cluster tasks. Secondly,the classified tasks were matched with classified resources to solve the low matching degree in resources and tasks. Experiments showed that the average execution time of tasks on the terminal was reduced by about 35. 2% after clustering task,and the task scheduling time had also been significantly reduced.