Study of scheduling algorithm based on T-LET planes for multiprocessor real-time systems

In this paper, a T-LET planes measure is applied to manage the task scheduling for multiprocessors. First, a novel scheduling algorithm on T-LET planes is proposed, and it is based on the strategy that the biggest M tasks, within the current remaining execution time, are first selected. Secondly, the algorithm has proved as an optimal multiprocessor scheduling algorithm for assigning tasks and currently feasible on T-LEF plane. Finally, this paper compares this algorithm with a global EDF algorithm by a specific example and experimental simulations. The simulation results show that the algorithm in this paper has the same complexity of time. Moreover, its effectiveness is higher and the scheduling success rate is higher.