Task allocation method based on heterogeneous multi-cores

Abstract

The invention discloses a task allocation method based on heterogeneous multi-cores. The task allocation method comprises the following steps: S1, construction of a task priority list: constructing the task list while scheduling task allocation; and S2, multi-strategy task allocation: inserting a task into a standby communication interval or directly copying a precursor of the task into a processor idle interval, allocating the task and a precursor node with the longest communication time to the same processor core, allocating a subsequent node with the longest task and the longest communication time to the same processor core, and selecting an optimal allocation scheme from the three. According to the invention, the allocation scheme with the shortest standby time of all processor cores can be effectively selected, the total execution time is reduced, and the high efficiency is ensured; the utilization rate of each processor core of a heterogeneous multi-core processor is improved, and as the communication time of high-priority tasks in the task execution process is shortened, energy originally consumed on the tasks is saved, and the heterogeneous multi-core processor has the energy-saving property.

Description

technical field [0001] The invention relates to the technical field of embedded scheduling of heterogeneous multi-cores, in particular to a task allocation method based on heterogeneous multi-cores. Background technique [0002] With the rapid development of computer technology in the 21st century, mankind has officially entered the information age. Computer technology has become more and more widely used in all walks of life, especially the rapid development of artificial intelligence technology in recent years. The development of artificial intelligence and machine learning technology has higher and higher requirements for computer hardware, especially the performance of computer CPU is particularly important. [0003] Since the structural design of each core of the homogeneous multi-core processor is the same, and there is no distinction in function and priority, it is impossible to increase efficiency by allocating program tasks on processor cores. Different application...

AI Technical Summary

Problems solved by technology

Each processor of a heterogeneous multi-core processor has a different structure, and different processors have different functions and scopes of application. Therefore, this processor is better at dealing with complex situations that require running multiple different programs.

But in this way, the task assignment and execution sequence of heterogeneous multi-core will be more complicated when actually executing tasks. Each application needs to find its corresponding optimal processor, and the difference in task volume will lead to the difference in the completion time of a single task.

[0005] The existing static heuristic scheduling algorithm has the problem of low processor utilization. The static heuristic algorithm has a lot of idle time when scheduling and assigning tasks to heterogeneous multi-core processors. Because the algorithm cannot properly allocate tasks to the corresponding The processor causes many processors to be idle after completing tasks, and the maximum efficiency of the processor is not achieved

Images