Energy consumption optimization method based on dynamic idle time mixed key periodic tasks

Abstract

The invention discloses an energy consumption optimization method based on a dynamic idle time mixed key period task, which comprises the following steps of: A, scheduling a mixed key period task setby using a dynamic priority strategy; b, calculating a unified static speed SU of the periodic task; c, recycling idle time IHI generated by the executed high-key-level periodic tasks and dynamic idletime IC generated when all the periodic tasks are completed in advance in the low mode; d, calculating the speed SL of the periodic task in the low mode according to the available idle time I = IHI +IC of the system, and when the execution time of the high-key-level periodic tasks in the low mode exceeds the worst execution time of the periodic tasks in the low mode, entering the high mode, executing the high-key-level periodic tasks in the high mode at the fastest processor speed all the time, and discarding all the low-key-level periodic tasks. The method has the advantages of reducing energy consumption and the like.

Description

technical field [0001] The invention relates to an energy consumption optimization method based on dynamic idle time mixed key cycle tasks. Background technique [0002] Real-time systems not only require logical correctness, but also have strict requirements on the time limit. With the development of real-time systems, they are widely used in production and life, especially in manufacturing, aerospace, medical and industrial production, etc.; and hybrid key systems are a typical application of real-time systems, which meet the requirements on the same platform. The requirements of different application levels, unmanned aerial vehicle and automobile autopilot system are typical applications; with the rapid development of processor technology, its energy consumption is getting higher and higher, high energy consumption not only affects the life of the processor and system reliability, but also Increase the production cost of products; energy consumption has become the main f...

AI Technical Summary

Problems solved by technology

[0003]Current research on hybrid critical systems focuses on the analysis of system feasibility and scheduling feasibility, while research on energy optimization methods for periodic task models of hybrid critical systems is relatively Only a few studies use fixed priority scheduling strategies, resulting in low system utilization, which cannot be applied to mixed critical systems with dynamic priorities; and these studies assume that periodic tasks are executed at their worst-case execution time, which 1. The assumption is too conservative and does not conform to the real situation of the system, resulting in excessive energy consumption of the system