An Improved Real-time Scheduling Algorithm Based on Deadline Constraint

Abstract

With multi-core processors becoming standard hardware platforms for various computing systems, the integration of real-time systems on multi-core hardware has become an inevitable trend. Designing real-time scheduling technology for multi-processor systems is a key issue for system designers. According to the urgent degree of tasks in different processing stages, an improved real-time scheduling algorithm based on deadline constraints (IDCSA) is proposed, including the task insertion feasible condition in the node queues and the task preemption condition of the node resources. While prioritizing urgent tasks, it avoids hanging tasks and waiting tasks missing their deadlines. The simulation results show that the algorithm has good load balancing performance and resource utilization. Compared with other two real-time scheduling algorithms, the success rate of task scheduling in IDCSA increases by more than 10% and 8% respectively.