A Game-Based Combinatorial Double Auction Model for Cloud Resource Allocation
Qihui Li, Chuanhe Huang, Haizhou Bao, Bin Fu, Xiaohua Jia · 2019
Cloud computing integrates a large number of resources through virtualization technology, and then provides users with personalized services on an on-demand basis. In response to this service model, this paper draws on the economic theories and proposed a game-based combinatorial double auction model for cloud resource allocation. Firstly, through Harsanyi transformation, the incomplete information game for cloud resource allocation is converted into a complete but imperfect information game, and the Bayesian Nash equilibrium solution is obtained. Then, we designed the resource allocation model supporting multiple infrastructure providers and service providers bidding on various combinations of resources. Considering both parties' interests, this model ensures fairness and high resource utilization. Simulation results show that this method not only forms a fair incentive mechanism for all parties in the transaction, but also optimizes the social welfare.