An Innovative Perspective on Mapping in Grids

To execute large scale applications exploiting the unemployed aggregated power available on grid nodes, effective and efficient mapping algorithms must be designed. Since the problem of optimally mapping is NP-complete, heuristic techniques can be profitably adopted to find near-optimal solutions. Here a multiobjective Differential Evolution algorithm is implemented and tested on different mapping scenarios with the aim to fulll several optimization criteria. The results attained show the robustness of the evolutionary approach proposed in dealing with multisite grid mapping.