Abstract：Resources scheduling is one of core technologies of grid computing，and it is a NP-complete problem．Genetic algorithm is search algorithms to solve task scheduling．An improved genetic algorithm is thus proposed for task scheduling on computational grid，and then the subtasks are ranked according to their depth value．This algorithm considering the algorithmic ability of computing resource，the relationship can be obtained through the DAG，solves wrong chromosome．This algorithm employ multi-group on evolution of population，strengthen the fatherly chromosome evolvement and recombined．The simulation comparison results show the effectiveness and ability of proposed algorithm to the scheduling problem is better than conventional genetic algorithm．
2010, Vol. 1 