欢迎您！东篱公司
退出
-
申报数据库
-
立项数据库
-
成果数据库
-
项目获奖数据库
中文摘要：
针对现有云计算任务调度研究忽视虚拟机间拓扑结构的问题,提出一种兼顾虚拟机资源拓扑结构的任务调度算法。首先,建立云计算任务模型和虚拟机资源拓扑模型,然后综合虚拟机处理性能和拓扑距离两个方面,提出虚拟机—任务适应度评价函数,计算、评价虚拟机的综合性能,将任务分配到综合性能优的虚拟机上。通过仿真实验将该算法与HEFT、DCP进行对比,实验结果表明,在考虑虚拟机间拓扑结构的情形下,面向不同类型、不同规模的任务集合,该算法比其他算法平均任务完成时间小,且具有较优的适应性。
英文摘要：
In the view of the issue that the existing studies of task scheduling didn＇t consider the topology of VMs in cloud computing, this paper proposed an algorithm of task scheduling taking VMs＇ topology into account. Firstly, the algorithm built models of cloud computing tasks and VMs＇ topology respectively. Then it introduced a VM-Task fitness evaluation function, in- tegrated with processing performance of VM and topological distance between VMs. After the evaluation of the function, it assigned the task to the VM with excellent comprehensive performance. The comparison experiments among the proposed algorithm, HEFT and DCP show that the proposed algorithm not only has good adaptability for different types of tasks or different scales of tasks, but also its average makespan is smaller than that of HEFT and DCP when considering VMs＇ topology.
同期刊论文项目
同项目期刊论文
Universal blind image quality assessment using contourlet transform and singular-value decomposition
期刊信息
位置：