基于动态能量感知的云计算任务调度模型

Cloud computing task scheduling model based on dynamic energy perception

摘要:

针对传统云计算任务调度模型出现的计算量大、能耗高、效率低、调配精度差等问题,基于动态能量感知设计了一种新的云计算任务调度模型。以动态能量感知为基础,选取资源分配服务器的中央处理器的使用率、存储器的占用率、控制器的负载率等三个参数,构建三维云计算任务节点投影空间,将上述参数向量投影到空间中。引入动态能量感知建立云计算任务调度模型,采用虚拟技术将多个服务器合并成一台服务器,对调度任务进行需求分析和分类,采用能量感知算法将待调度任务分配给满足调度需求的虚拟资源,将任务调度到服务器资源上,实现任务调度。实验结果表明,基于动态能量感知的云计算任务调度模型在从小任务集和大任务集两个角度都能给有效缩短调度时间,降低调度能耗。

Abstract:

In view of the large amount of calculation, high energy consumption, low efficiency, and poor deployment accuracy of traditional cloud computing task scheduling models, a new cloud computing task scheduling model is designed based on dynamic energy perception. Based on dynamic energy perception, three parameters such as the utilization rate of the central processing unit of the resource allocation server, the occupancy rate of the memory, and the load rate of the controller are selected to construct a three-dimensional cloud computing task node projection space, and the above parameter vectors are projected into the space middle. Introduce dynamic energy perception to establish a cloud computing task scheduling model, use virtual technology to merge multiple servers into one server, analyze and classify scheduling tasks, and use energy perception algorithms to assign tasks to be scheduled to virtual resources that meet scheduling requirements. Tasks are scheduled to server resources to achieve task scheduling. The experimental results show that the cloud computing task scheduling model based on dynamic energy perception can effectively shorten the scheduling time and reduce the scheduling energy consumption from both the small task set and the large task set.