A Resource Allocation Model with Cost-Performance Ratio in Data Grid

Resource allocation for data transfer is a fundamental issue for achieving high performance in data grid environments. In this paper, we survey the existing researches on allocation problems in grid environment and propose a resource allocation model based on the cost-performance ratio for commerce environments. This ratio takes both price and quality of data resource into consideration at the same time. According to the optimization objective we define two types of ratios: Price-aware ratio and quality-aware ratio. They are suitable for the environments where allocations put more emphasis on quality or price of resource respectively. Based on the maximal cost-efficiency, we formulize the allocation optimization problem and present two theorems and a deduction about its solutions. The corresponding proofs are also given in the paper. Finally, we present two algorithms to allocate and re-allocate data resources. The analysis shows that the algorithms can satisfy our requirements.