Energy-efficient mixed task scheduling algorithm on heterogeneous multiprocessors

Lei Chen; Hao Lan; Xiangzhan Wu

doi:10.1117/12.2639179

29 April 2022 Energy-efficient mixed task scheduling algorithm on heterogeneous multiprocessors

Lei Chen, Hao Lan, Xiangzhan Wu

Proceedings Volume 12255, 2022 International Conference on Optoelectronic Information and Functional Materials (OIFM 2022); 1225516 (2022) https://doi.org/10.1117/12.2639179
Event: 2022 International Conference on Optoelectronic Information and Functional Materials (OIFM 2022), 2022, Chongqing, China

Abstract

To solve the high energy consumption problem for heterogeneous multiprocessors, an energy efficient mixed task scheduling algorithm (EMA) for periodic and aperiodic tasks is presented. The algorithm assigns tasks to specified core according to their impact on system energy consumption and average worst-case execution time (WCET). At the same time, the sum of periodic task utilization and the bandwidth of the total bandwidth server on a single core less than 1 to ensure the scheduling scheme feasibility. In the core with only periodic tasks, let WCET close to the deadline, and reduce the execution frequency to decrease the energy consumption. On the core with mixed tasks, tasks were scheduled by the earlier deadline first and total bandwidth server. When aperiodic tasks are executed, the processor works at the maximum executable frequency to ensure timely response. Simulation results show that the energy consumption of EMA is 46.7%~69.0% lower than that of the typical heuristic, the execution time is much less than that of the algorithm based on Optimization theory, and the scheduling scheme is more feasible and practical.

Citation Download Citation

Lei Chen, Hao Lan, and Xiangzhan Wu "Energy-efficient mixed task scheduling algorithm on heterogeneous multiprocessors", Proc. SPIE 12255, 2022 International Conference on Optoelectronic Information and Functional Materials (OIFM 2022), 1225516 (29 April 2022); https://doi.org/10.1117/12.2639179

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available