Adaptive resource discovery models and Resource Selection in grids

Resource Discovery and Resource Selection are the crucial tasks in grid scheduling and resource management. The goal of resource discovery and selection is to identify list of authenticated resources that are available in the grid for job submission and to choose the best node. Recently adaptive push-pull method was proposed for information dissemination of information for dynamic web data. In this paper, we extend the ideas of the adaptive push-pull method to grid environment for resource discovery. We also perform a comparison of the adaptive push-pull method with adaptive pull and push resource discovery models. The experiments are performed by varying CPU and RAM weights to investigate the impact of the selection of these weights on resource discovery. From our experiments Adaptive push-pull method outperformed other two models in terms of selecting a unique node for job submission. It is also clear that resource evaluation index increases as the CPU weight increases.