Research On Resource Scheduling Algorithm And Monitor System Implementation Based On Cloud Platform

The emergence of cloud computing, makes computing resource used by ordinary users like water and electricity power. There is no doubt that many enterprises will get huge benefits from it. Many IT companies developed their own cloud computing platform, such as Amazon EC2, Google’s App Engine, as well as Microsoft’s Azure and so on. But computing resource utilization efficiency of these platforms are not high and load imbalance is serious, the key to solving these problems lies in the application of efficient resource scheduling algorithm of cloud computing platform. So this paper focuses on the resource scheduling problems of cloud computing environment.The main work of this paper has following aspects:(1) For resource scheduling problem in cloud computing platform, with an in-depth analysis of composite principle of cloud platform, constructs a cloud system model structure, and researches associated resource scheduling algorithm based on this structure.(2) For system resource utilization is not high and load imbalance problems in cloud computing platform, we propose a dynamic resource scheduling based on weighted maximize integrated resource utilization efficiency. We use cloud simulation platform Cloud Sim to have a test. By comparing with Myopic Maxweight algorithm and Random algorithm, indentifies the effectiveness of the proposed algorithm.(3) For load imbalance problem in cloud computing platform during the running process, we propose a virtual machine live migrate algorithm based on minimum load imbalance. And the algorithm is verified through simulation, the results show that the algorithm can achieve high resource utilization, while the degree of load imbalance can be maintained at a lower level.(4) Considering the open source cloud platform Cloud Stack can’t effectively monitor virtual machine resource utilization, this paper presents a virtual machine monitored method based on Cloud Stack cloud platform. Through a deployment and test, result verifies the feasibility of the method.