GRASP Algorithm for Optimization of Grids for Multiple Classifier System

摘要

In recent years the volume of data used in scientific researches and industry has increased significantly. Distributed computing systems including Grids use the public Internet to share computational resources of research institutions around the world in order to process the data. Due to large data volumes being transferred, network aspects of Grids have become important. In this work we introduce a model of an overlay Grid system, which could be used by the distributed recognition system based on the idea of combining classifiers. We formulate an Integer Programming optimization problem with the objective to minimize the overall cost including processing and data transfer. Next, an effective heuristic algorithm is developed to solve the problem. Results of numerical experiments showing the comparison of the heuristic against solutions provided by CPLEX solver are presented.