Hybrid task scheduling method for heterogeneous multi-core processor

Abstract

The invention provides a hybrid task scheduling method for a heterogeneous multi-core processor. According to the method, optimization is carried out on the basis of a sparrow search algorithm, in task scheduling in a heterogeneous multi-core environment, priority levels of task nodes in an HEFT algorithm are sorted, a task scheduling list is constructed, meanwhile, a reasonable task allocation coding scheme is designed, sparrow search space is mapped to discrete space, and therefore, the sparrow search algorithm is suitable for the research of discrete heterogeneous multi-core task schedulingproblems. According to the method, the HEFT algorithm and the sparrow search algorithm are mixed, the task list obtained by the HEFT algorithm is added into the initial population of the sparrow search algorithm, iteration of the algorithm is executed by utilizing the advantages of high optimization capability, high convergence speed, stable performance and the like of the sparrow search algorithm, and the task with the highest priority is taken out from the list, and is allocated to the processing core with the earliest starting time. The task execution time is effectively shortened, and thetask scheduling efficiency in the heterogeneous multi-core environment is improved.

Description

technical field [0001] The invention relates to the field of operating systems, in particular to a task scheduling method in which a table scheduling algorithm and a sparrow search algorithm are mixed. Background technique [0002] With the development of high-performance computing and the rise of technologies such as AI, machine learning, and cloud technology, a processor system with a single structure cannot simultaneously meet the diverse needs of parallel computing of applications. Therefore, heterogeneous multi-core processors came into being and gradually entered the market. Heterogeneous multi-core processors are asymmetric multi-core processors composed of processing cores of various types and different computing capabilities. How to coordinate the work among these processing cores, divide tasks reasonably, and assign them to processors for execution, so as to improve the performance of heterogeneous multi-core processors. The efficiency of task scheduling in the en...

AI Technical Summary

Problems solved by technology

However, the genetic algorithm is complex in operation and poor in real-time performance, and it is difficult to guarantee the optimization time; although the particle swarm algorithm is simple in operation and easy to implement, it is prone to the predicament of particles falling into "premature" phenomenon and local optimum in the later stage.

Images