A Task Priority Design Method Based on Multi-Feature Parameters in Real-Time Scheduling
LI Qing-hua
Abstract
LI Qing-hua
Abstract
This paper proposes a task priority design method based on four feature parameters:task value,remaining execution time,slack time and arrival time,and makes the task priority change accordingly with the change of the task's degree of urgency and the percentage of the completion of the task.A dynamic preemptive scheduling algorithm is presented based on the new task priority design method,which is called VRSAF.The simulation results show that the scheduling performance of VRSAF is close to EDF algorithms in light load situation.And in overload situation,the performance of scheduling is higher than the HVF algorithm.The overall activation performance is high,especially it can degrade gracefully under the overload situation.
OpenAlex reports 1 citations for this work. Citation counts describe recorded attention and do not establish research quality.
A contribution statement is not available in the OpenAlex record.
Method details are not available in the OpenAlex metadata.
Findings are not separately available in the OpenAlex metadata.
Limitations are not available in the OpenAlex metadata.
Application details are not available in the OpenAlex metadata.
This paper proposes a task priority design method based on four feature parameters:task value,remaining execution time,slack time and arrival time,and makes the task priority change accordingly with the change of the task's degree of urgency and the percentage of the completion of the task.A dynamic preemptive scheduling algorithm is presented based on the new task priority design method,which is called VRSAF.The simulation results show that the scheduling performance of VRSAF is close to EDF algorithms in light load situation.And in overload situation,the performance of scheduling is higher than the HVF algorithm.The overall activation performance is high,especially it can degrade gracefully under the overload situation.
Key concepts: Computer science, Priority inversion, Deadline-monotonic scheduling, Priority inheritance, Earliest deadline first scheduling, Scheduling (production processes), Dynamic priority scheduling, Fixed-priority pre-emptive scheduling