摘要
云环境中如何为任务分配虚拟资源并将其调度到物理资源上是一个研究难点;在详细分析云环境中任务分发流程的基础上,构建了虚拟资源的调度控制模型,并提出采用分布估计算法(Estimation of distribution algorithms,EDAs)进行求解;该模型首先通过感知器感知物理资源,然后将物理资源和虚拟资源抽象为具有一定属性的节点,资源的分配过程转化为将虚拟资源映射到物理资源;同时提出了资源满足率的概念并以此为目标函数来进行优化,对比Max-min算法,静态调度算法和随机调度算法,在任务集为5~55的区间及负载量为0.5~1.5的区间,得出EDA算法的资源满足率平均至少提高了1.004倍,最高达1.793倍。
It's a research difficulty how to allocate virtual resources to physical resources for tasks and schedule virtual resources to physical resources in cloud environment. This paper constructed a virtual resources scheduling model and proposed using Estimation of Distribution Algorithms (EDAs) to resolve it based on the detail analyzing the flow of task assignment. The physical resources were perceived through perceptron firstly in this model, and then the physical resources and virtual resources were abstracted to some nodes with attributes, the process of resources assignment was changed to mapping virtual resource to physical resource. At the same time, the concept of rate of resource service was proposed as the optimum function. When the tasks number is between 5 and 55 and the load rate is between 0. 5 and 1.5, comparing with Max--min algorithm, static algorithm and random algorithm, the rate of resource service of EDA algorithm is improved on average by at least 1. 004 and at most 1. 793 times.
出处
《计算机测量与控制》
CSCD
北大核心
2011年第12期3145-3148,共4页
Computer Measurement &Control
基金
浙江省高职高专院校特色专业建设项目(TZZ09085)
关键词
云计算
资源调度
分布估计算法
Cloud computing
resources scheduling
Estimation of distribution algorithms