Heterogeneous multi-core energy-saving task schedule method based on improved genetic algorithm

Abstract

A heterogeneous multi-core energy-saving task schedule method based on improved genetic algorithm comprises the improved genetic algorithm used for determining task priority and energy-saving schedule algorithm based on zooming priority. The flow of the heterogeneous multi-core energy-saving task schedule method includes (1), initializing population information; (2), entering a loop body and determining the task priority by the aid of genetic algorithm; (3), determining schedule sequence of tasks on a processor according to a task DAG (directed acyclic graph) and division strategies; (4), realizing dynamic voltage zooming on the basis of feasible task schedule according to relation of saving energy of the tasks and prolonging time; (5), calculating fitness of a current population and sorting the current population; and (6), updating the population by the aid of the improved genetic algorithm, determining new task priority, quitting if termination conditions are met, and continuing iteration if the termination conditions are not met.

Description

technical field [0001] The invention mainly relates to the design field of embedded systems, in particular to a heterogeneous multi-core energy-saving task scheduling method based on an improved genetic algorithm. Background technique [0002] In order to solve the power consumption problem in embedded system design, the dynamic voltage scaling technology emerging in recent years has attracted extensive attention in the industry. At present, many popular low-power processors support DVS (Dynamic Voltage Scaling) technology, which reduces system power consumption by changing the voltage and frequency of the processor unit in real time. However, the reduction of system frequency will lead to longer task execution time, which may affect the real-time performance of the system. Therefore, in the process of voltage scaling, it is necessary to select an appropriate scheduling method to optimize power consumption without affecting system performance. [0003] Energy-efficient sch...

AI Technical Summary

Problems solved by technology

[0004] In the prior art, some formulate the voltage scaling problem as an integer programming problem, and the representative one is an energy-aware algorithm that simultaneously schedules multi-rate periodic tasks and non-periodic tasks, but the complexity of the algorithm based on integer programming is too large, Difficult to be widely used in practice

[0005] Some researchers combined stochastic algorithms and energy gradient techniques to simultaneously solve the allocation of slack time slices and task scheduling. However, the algorithm for evenly or randomly allocating slack time slices ignores the energy-saving differences of different tasks and cannot achieve the optimal energy-saving effect.

[0006] There is also a priority based on the energy gradient and execution time of the task, and then random scheduling under the guidance of the priority and a priority-based list scheduling algorithm, but the purely priority-based scheduling algorithm has great limitations in the solution space property, it is difficult to obtain the optimal solution of the system

[0007] In addition, the traditional genetic algorithm and list scheduling strategy are combined, and the genetic list scheduling algorithm is proposed to determine the execution order of tasks, and the related dynamic voltage scaling algorithm is combined to complete the energy-saving scheduling in the heterogeneous multi-core environment, but the scheduling algorithm based on the traditional genetic algorithm Although it has a wide solution space, the local optimization function is not strong, which affects the energy saving effect

Images