基于卡尔曼预测器的云计算资源调度研究

摘要

Resources scheduling is one of the main research directions of cloud computing, in order to improve the utilization rate of cloud resource scheduling and keep load balance. Research on the existing cloud computing resource scheduling algorithm, virtual resource scheduling model in cloud computing is provided and a cloud computing re-source scheduling algorithm based on KalmanPredictor is designed. The Kalman Predictor is used to predict the state of the virtual resources to obtain the resources vector. Then the calculation task of cost and building the mapping model of the tasks and resources are conducted. The task is assigned to the appropriate virtual resource scheduling to avoid the blind scheduling and the experiment simulated onCloudSim platform shows that this algorithm can effectively solve theResource Schedulingin cloud computing with a smaller load balancing deviation and task make span. It has larger priority than other methods.%资源调度是云计算的一个主要研究方向，本文在研究现有云计算资源调度算法的基础上，为了提高云计算资源的利用率，保持负载平衡，设计了基于卡尔曼预测器的云计算资源调度算法，采用卡尔曼预测器预测虚拟资源下一时刻的状态，由此得到虚拟资源的向量，然后计算任务的代价，建立任务与资源的映射模型，将任务分配到合适的虚拟资源进行调度，避免了盲目调度，提出了基于卡尔曼预测器的云计算资源动态调度的具体算法，最后在云计算仿真平台CloudSim上进行仿真实验与分析。仿真实验的结果证明，本文提出的算法在任务执行的跨度和负载均衡离差方面具有一定的优势。