云资源分配优化的非合作博弈定价机制

Pricing Mechanism Based on Non-cooperative Game for Optimizing Cloud Resource Allocation

云资源提供者向用户提供服务,要争取更多用户,服务所需资源的定价就至关重要。以在满足用户需求的同时达到最大化收益为目标,提出了一种在云计算环境中基于非合作博弈模型的资源定价机制。在博弈模型中,用户效用模型对应定价的资源需求函数,云提供者收益模型对应需求的资源价格函数,根据需求提供不同价格资源的云提供者和依据所提价格更新需求的用户组成博弈结构,博弈Nash均衡定义为供求双方决定的最优定价。从理论上证明了所提博弈机制存在资源定价的Nash均衡解且是唯一的,仿真实验验证算法是有效可行的。

To compete for cloud users,it is critically important for each cloud resource provider to select an optimal price which is best corresponds to their service qualities,to remain attractive to cloud users.In this paper,the resource pricing problem is defined as the non-cooperative game model in cloud computing.Our objective of this game is to enable cloud providers to maximize their profit while satisfying cloud users by maximizing their utility.In this game model,the utility of users is modeled as a function of resource demand with a corresponding price,and the profit of the cloud provider is modeled as a function of resource price with a corresponding demand from users.The game consists of the cloud provider suggesting differentiated resource prices according to the demand and users updating their requests in view of the proposed price.The Nash equilibrium solution of our game is defined as the optimal suggested prices by the cloud provider and the optimal user demands.It proves that the proposed game model has Nash equilibrium of resrouce pricing,and this Nash equlibrium solution is unique.The experimental results show that our game algorithm performs better.