Time slice-based task scheduling method

Abstract

The invention discloses a time slice-based task scheduling method. The realization method comprises the steps of: when a system detects a newly arrived task or task completion, performing re-initialization on task parameters of all tasks immediately; then splitting each task into sub tasks with reasonable execution time slice lengths; determining an execution sequence of the sub tasks and a processor for executing the sub tasks; and finally, updating a task scheduling list, scheduling task execution according to a new scheduling list, and feeding back task completion information to the abovementioned task parameter initiation part in time so as to promote to generate a new task scheduling list. The time slice-based task scheduling method can be used for an isomorphic multi-processor system.

Description

technical field [0001] The invention belongs to the field of operating systems, and in particular relates to a task scheduling method based on time segments, which can be used in homogeneous multiprocessor systems. Background technique [0002] The task scheduling method is a key issue that must be considered when designing a high-performance multiprocessor system. With the aggravation of global energy shortage, how to reduce processor energy consumption has attracted more and more attention. Therefore, a good task scheduling method can make the system run at the lowest rated power as possible while ensuring the quality of service, so as to achieve the purpose of energy saving. [0003] Existing background technology usually adopts such as figure 1 The open-loop scheduling method shown. When multiple tasks enter a multiprocessor system, the system estimates an execution time for each task based on the size of the individual tasks. When a task executes on a processor, it ...

AI Technical Summary

Problems solved by technology

Since the estimated execution time is given based on the worst case, the task is often completed earlier than the estimated execution time, and the rest of the time, even if there is no task running on the processor, it still needs to be completed at the previous time. When the running frequency of the task is executed, at this time, not only the remaining time is wasted, but also unnecessary energy consumption will be generated.

[0004] The greater the difference between the estimated execution time of the task and the actual execution time, the greater the uncertainty of the system environment and the greater the energy wasted by the system

Therefore, it is necessary to find a task scheduling method to overcome the energy waste caused by the uncertainty of the system environment

Images