An Energy Saving Scheduling Method Applicable to Periodic Tasks of Numerical Control System

Abstract

The invention discloses an energy-saving scheduling method suitable for numerical control system periodic tasks. The method comprises the following steps: calculating the optimal operating speed S of each task Ti under an off-line state in a numerical control system, then calculating the execution time of the task Ti at the optimal operating speed S, and sorting a task set according to an earliest deadline first principle; when a certain task is finished, recovering the slack time Slack_Time of the task Ti, and using the slack time Slack_Time to calculate the operating speed S of the task under a specified continuous voltage of a processor; finally determining the preceding-part operating speed SL and the later-part operating speed SH of each task according to the operating speed S, so as to calculate the execution time ex.L of each task at the preceding-part operating speed SL and the execution time ex.H at the later-part operating speed SL, and scheduling the execution tasks according to the calculated real execution time, i.e. the ex.L and the ex.H. In such a way, the slack time Slack_Time of a system is fully utilized, the operating speed of the processor is reduced, and the aim of energy consumption reduction is achieved.

Description

technical field [0001] The invention relates to real-time scheduling of tasks in the field of real-time systems, in particular to an energy-saving scheduling method suitable for periodical tasks of a numerical control system. Background technique [0002] The real-time system is more and more widely used in the field of numerical control technology and embedded products, especially in the applications with complex functions and huge systems. This requires real-time systems to provide more efficient computing power to meet the requirements of wireless communication and multimedia applications, but the cost of high performance is high energy consumption. At the same time, with the increasingly sophisticated manufacturing process and the rapid development of the scale of integrated circuits, the power consumption of the system has increased sharply. The high temperature caused by high power consumption will increase the possibility of system failure and reduce the reliability o...

AI Technical Summary

Problems solved by technology

[0004] The existing low-power scheduling algorithm adopts DVS technology. On the premise of satisfying the real-time performance of the system, by reclaiming the idle time of the system, the operating voltage and clock frequency of the processor are reduced by using the idle time to reduce the system energy consumption; but its assumption The processor provides continuous frequency and voltage, while the actual commercial processor only provides discrete frequency and voltage. In order to ensure that the task does not miss the deadline, the actual operating speed is selected to be greater than the ideal speed calculated under the continuous frequency case, so that Will cause waste of system resources

Images