Energy and performance trade-off optimization in heterogeneous computing via reinforcement learning

Yu, Z., Machado, P., Zahid, A., Abdulghani, A. M., Dashtipour, K., Heidari, H. , Imran, M. A. and Abbasi, Q. H. (2020) Energy and performance trade-off optimization in heterogeneous computing via reinforcement learning. Electronics, 9(11), 1812. (doi: 10.3390/electronics9111812)

[img] Text
225757.pdf - Published Version
Available under License Creative Commons Attribution.

643kB

Abstract

This paper suggests an optimisation approach in heterogeneous computing systems to balance energy power consumption and efficiency. The work proposes a power measurement utility for a reinforcement learning (PMU-RL) algorithm to dynamically adjust the resource utilisation of heterogeneous platforms in order to minimise power consumption. A reinforcement learning (RL) technique is applied to analyse and optimise the resource utilisation of field programmable gate array (FPGA) control state capabilities, which is built for a simulation environment with a Xilinx ZYNQ multi-processor systems-on-chip (MPSoC) board. In this study, the balance operation mode for improving power consumption and performance is established to dynamically change the programmable logic (PL) end work state. It is based on an RL algorithm that can quickly discover the optimization effect of PL on different workloads to improve energy efficiency. The results demonstrate a substantial reduction of 18% in energy consumption without affecting the application’s performance. Thus, the proposed PMU-RL technique has the potential to be considered for other heterogeneous computing platforms.

Item Type:Articles
Status:Published
Refereed:Yes
Glasgow Author(s) Enlighten ID:Abbasi, Professor Qammer and Imran, Professor Muhammad and Yu, Zheqi and Zahid, Mr Adnan and Heidari, Professor Hadi and Abdulghani, Dr Amir Mohamed and Dashtipour, Dr Kia
Creator Roles:
Yu, Z.Conceptualization, Methodology, Software, Validation, Formal analysis, Investigation, Resources, Data curation, Writing – original draft, Visualization
Zahid, A.Writing – review and editing
Abdulghani, A. M.Writing – review and editing
Dashtipour, K.Writing – review and editing
Heidari, H.Supervision
Imran, M. A.Supervision
Abbasi, Q. H.Supervision, Project administration, Funding acquisition
Authors: Yu, Z., Machado, P., Zahid, A., Abdulghani, A. M., Dashtipour, K., Heidari, H., Imran, M. A., and Abbasi, Q. H.
College/School:College of Science and Engineering > School of Engineering > Electronics and Nanoscale Engineering
College of Science and Engineering > School of Engineering > Systems Power and Energy
Journal Name:Electronics
Publisher:MDPI
ISSN:0013-5070
ISSN (Online):0883-4989
Published Online:02 November 2020
Copyright Holders:Copyright © 2020 The Authors
First Published:First published in Electronics 9(11): 1812
Publisher Policy:Reproduced under a Creative Commons License

University Staff: Request a correction | Enlighten Editors: Update this record

Project CodeAward NoProject NamePrincipal InvestigatorFunder's NameFunder RefLead Dept
172865EPSRC DTP 16/17 and 17/18Tania GalabovaEngineering and Physical Sciences Research Council (EPSRC)EP/N509668/1Research and Innovation Services