Mechanism Design And Game Analysis Of Resources Allocation In Cloud Computing

Cloud computing is an emerging model of computing, it is a new method of shared infrastructure to provide resources to cloud users in a large scale distributed environment, such as computing resources, data storage and network services. By using cloud computing, cloud users(enterprise or individual) can not only reduce the cost, but also can focus on improving their core competitiveness and not have to worry about whether their own infrastructure can support. With the development of the could computing, resources allocation and management has become a critical issue in cloud computing. The allocation of cloud resources in a reasonable mechanism can not only improve the utilization of cloud resources, but also can meet a variety of users’ demands, making the market of cloud resources services more stable and orderly.In the traditional allocation of cloud resources, cloud users buy or adjust its service resources according to the actual demands and pay the fee as their usage, which is the initial business model of could computing. With the development of the cloud computing business and the needs of users become diversity day by day, cloud providers have started to try to create new allocation mechanism of cloud resources, such as based on resource reservation mechanism and online auction mechanism etc, the users’ demands are taken more consideration in these new mechanisms. This paper assumes that each cloud user is rational and intelligent, establishes the related allocation models of cloud resource, and study the allocation mechanism problems of cloud resource by using game theory, mechanism design and relate theory. The work content and innovation are as follows:(1) The traditional allocation mechanism of cloud resources is on-demand according to the users’ demands, combined with a hybrid cloud environment, the paper builds a workload factoring game model, and analyzes strategies of users in hybrid cloud according to the performance analysis and market analysis. The paper gives the proof of existence and uniqueness of Nash equilibrium in performance analysis, and gives an algorithm to obtain the Nash equilibrium. Through the market analysis, the Nash equilibrium expression is obtained, the paper makes analysis of the Nash equilibrium strategies under some data.(2) The resources requests of cloud users are dynamic, the paper designs an online mechanism from the cloud provider perspective to allocate cloud resources by bidding form, which is different from traditional allocation mechanism of cloud resources. The online mechanism can meet individualized demands of users, the incentive compatible in this mechanism can be satisfied simultaneously, and the mechanism is compared with the traditional offline mechanism, simulation analysis shows the mechanism has a good performance.At present, cloud computing is developing to the direction of diversification, at the lack of research on the allocation mechanism of cloud resources, the results of the study in this paper has a certain reference value to national development and cloud computing services.