Dynamic scheduling algorithm for parallel real-time jobs in heterogeneous system

The dynamic tasks scheduling of parallel real-time jobs in heterogeneous system is still challenging problem that is being investigated by some researchers. But the DAG-based real-time tasks scheduling isn´t yet paid enough attention. This paper proposed DAG-based real-time tasks scheduling model and a real-time scheduling algorithm DEFF that has a lower time complexity. Furthermore, the simulation experiments show that the scheduling model and scheduling algorithm are feasible, a higher scheduling successful ratio may be obtained by this algorithm under the condition of a small or middle scale of parallel jobs.