The Research And Application Of The Key Technologies Of Cloud Computing Management Platform Based On Cloudstack

In the cloud computing era, users can customize service, applications and resources according to their requirements and preferences. Cloud computing platform can deploy the service, resources, computing ability and applications according to users’requirements. In the cloud environment, the virtual machine is the large size of resources. Users usually require its operation stability, and don’t want to see the phenomenon of frequent migration of virtual machines. If we want users to use the resources in the cloud environment just like using the water and electricity, then handling the resource allocation is the key issue.Resource scheduling has two objectives:(1) load balancing.(2) energy-saving. The so-called load balancing means to ensure that the load rate of hosts on cloud platform is balanced, and there’re no hosts whose load is too heavy or too light. Energy-saving, is to try to reduce the number of hosts on the cloud platform, turn off some hosts whose load is too light. In this way, we can achieve the goal of energy conservation. However, general virtual machine’s deployment strategy is difficult to balance between these two scheduling goals. This paper presents an optimized deployment scheme of virtual machine based on the individual requirements of users. Firstly, this scheme gives weights to the four types of resources (CPU, memory, hard disk and bandwidth) according to the different application background of users’requirements. Then, select a candidate target host by comparing the resources values which user requests with the available resources values of hosts in the cloud platform. Finally, calculate the objective function value of each candidate target host, then choose the host which has the maximum objective function value as the target host to deploy the virtual machines.Experimental results show that this scheme can select the optimal host to deploy the virtual machine. It not only can use resources scientifically and rationally, but also can improve the phenomenon of load imbalance effectively. What’s more, it can obtain a better users’experience. Considering about the characteristics of demand charges of cloud computing, this paper did a research about the billing strategy of Amazon Web Service, and developed a billing system based on CloudStack cloud management platform.