Hybrid game based virtual resource pricing model in cloud environment

Peng Xiao,Zhigang Hu
DOI: https://doi.org/10.13196/j.cims.2014.01.xiaopeng.0198.9.20140125
2014-01-01
Abstract:To solve the deficiency of existing pricing mechanism in aspects of efficiency and fairness, a cloud resource pricing model based on hybrid game strategy was proposed. In this model, the configuration and provision of virtual resource were transformed into a two-phrase gaming procedure. In first stage, the cloud resource benefits was optimized by using cooperative gaming strategy; in second stage, the user costs and cloud resource benefits were balanced by using non-cooperative gaming strategy. Through the theoretical analysis, the validity and solution of the proposed model were presented, and the experimental results indicated that the hybrid gaming model could significantly improve the price negotiation efficiency and reduce the application execution latency caused by conventional price negotiation mechanisms. For the task load with high concentration ratio, the resource selection and the matching strategy of the proposed model showed better load sensing ability and fairness, which could improve the Quality of Service (QoS) satisfaction of tasks in resource cost and deadline guarantee.
What problem does this paper attempt to address?