Optimizing Grid Resource Combination Using Frequent Pattern Mining

While most of the existing grid literatures only consider the resource pricing strategy, this paper focuses on optimizing grid resource combination, another important issue the grid service providers should address to maximize their profits. We develop an efficient algorithm for computing and maintaining all the frequent demand patterns and dynamically updating them with the incoming grid trade data stream. In our design, we introduce a support framework with weighted grid trade data and actively maintain pattern frequency histories under a summary table structure. Moreover, this paper shows the grid resource combination adjusting strategy for the grid service providers. Simulation results prove the well performance of the proposed method