Optimization of Grid Resource Allocation Using Improved Particle Swarm Optimization Algorithm

To solve the problem of grid resource allocation for tasks, an allocation algorithm based on improved particle swarm optimization was proposed. This algorithm leaded the cross operation, variation operation and select operation of the GA to the Particle Swarm Optimization Algorithm, it effectively overcame the inherent flaw of getting local optimal value by particle swarm algorithm and find the global optimum value in the search space again. The method is simple, needs less parameters, easy to programme, and ensures that particles in the update process control in integer space, avoiding unnecessary rounding of real numbers, and into local optimum problem, speeds up the convergence rate. After searching of particle in each sub-swarm, an optimal scenario for grid resource allocation was produced. Simulation experiments demonstrated effectivness and feasibility of the algorithm and achieves a better result in the aspect of grid resource allocation.