Low-energy-consumption task scheduling strategy for CPU-GPU heterogeneity

Abstract

The invention discloses a low-energy-consumption task scheduling strategy for CPU-GPU heterogeneity, and provides an ant colony task scheduling algorithm capable of simultaneously paying attention toreal-time constraints and system energy consumption aiming at the characteristics of a heterogeneous multi-core system and the problems that a traditional ant colony algorithm only optimizes a singletarget and is too slow in convergence speed. The method comprises the following steps: firstly, providing guidance information in a pheromone initialization process according to the energy consumptionof a task on a heterogeneous core; accelerating algorithm convergence speed, then, the cores are screened through task real-time constraint conditions; the method comprises the steps of obtaining a heterogeneous core of a task, selecting an appropriate execution core according to the calculation energy consumption of the task on the heterogeneous core, the inter-core communication energy consumption of different tasks and the pheromone content, continuously searching a scheduling scheme with lower energy consumption through multiple iterations of an ant colony algorithm, and adjusting the pheromone content according to the obtained result to further accelerate the convergence speed of the algorithm. A final task scheduling scheme is obtained after a plurality of iterations, so that the energy consumption of the system can be optimized under the condition that the real-time constraint of the task is met.

Description

technical field [0001] The invention belongs to the field of heterogeneous system task allocation and resource scheduling, and specifically relates to a resource allocation and task scheduling strategy oriented to CPU-GPU heterogeneous system structure while considering real-time constraints and system energy consumption. Background technique [0002] Nowadays, the processor architecture is gradually developing towards multi-core and heterogeneous. Multi-core processors have become the current mainstream processors. With multiple processing cores, the application program can deliver parallel threads to multiple cores for processing separately, so the running speed of the program is greatly improved. According to whether there are differences in their core structures, multi-core processors are divided into homogeneous multi-core processors and heterogeneous multi-core processors. A homogeneous multi-core processor mostly means that all processor cores have the same architec...

AI Technical Summary

Problems solved by technology

Zhang Jing et al. proposed a real-time task scheduling algorithm suitable for heterogeneous multi-core systems by remodeling the heterogeneous core according to the processing capability and scope of application, and referring to the main idea of ​​the relatively strict task scheduling algorithm. It has indeed achieved good results in terms of real-time performance, but the scope of application of processing tasks is still limited by high-real-time processing methods, which is not flexible enough

Bai Enci and others also proposed a heterogeneous multi-core task scheduling algorithm based on ant colony algorithm, and also considered real-time constraints when improving ant colony algorithm, and initially combined real-time constraints with performance-oriented ant colony algorithm up, but did not take into account the soft real-time requirements and the lack of information in the initial stage of the ant colony algorithm

[0005] To sum up, although these improved task and resource scheduling strategies can be applied to heterogeneous multi-core, and some real-time considerations are considered, the distinction between heterogeneous cores is still not detailed enough, and most of them are aimed at improving performance. It does not consider that the ant colony algorithm can be used to search for a feasible solution that satisfies the real-time constraints and no longer optimizes the task scheduling length, but optimizes the overall energy consumption.

Images