A Game-Theoretic Resource Allocation Strategy with Purification Approach for Computational Grids

Under the context of grid environment, a methodology to define a sustainable, market-like resources allocation using economic and technical criteria is proposed. Based on the associated theory of sales, the resource allocation strategy is identified to address resource sharing and collaboration, which is economically efficient and necessary to allow the development of market instruments and the supply of demand. Scrutinizing the grid by taking into account an explicit purification of the mixed-strategy equilibrium, the model of cooperation and interaction between grid sites is achieved. In addition, the benefit from the respective grid system at a minimum cost and in an efficient manner is identified. The results show that the proposed approach has advantage over other scheduling schemes in optimizing incentives for both consumers and providers, leading to highly successful job execution and fair profit allocation.