task scheduling using particle swarm optimization algorithm with a selection guide and a measure of uniformity for computational grids

in this paper, we proposed an algorithm for solving the problem of task scheduling using particle swarm optimization algorithm, with changes in the selection and removing the guide and also using the technique to get away from the bad, to move away from local extreme and diversity. scheduling algorithms play an important role in grid computing, parallel tasks scheduling and sending them to appropriate resources. the proposed method has less makespan and price. in addition to implementing a grid computing system, the proposed method which is using three standard test functions in evolutionary multi-objective optimization is evaluated. in this paper, the number of elements in the assessment of the pareto optimizes set, uniformity and error. the results show that this search method has more optimization in particle number density and high accuracy with less error than the mopso and can be replaced as an effective solution for solving multi-objective optimization.