King's College London

Research portal

Collaborative Adaptation for Energy-Efficient Heterogeneous Mobile SoCs

Research output: Contribution to journalArticle

Standard

Collaborative Adaptation for Energy-Efficient Heterogeneous Mobile SoCs. / Singh, A.K.; Basireddy, K.R.; Prakash, A.; Merrett, G.V.; Al-Hashimi, Bashir M.

In: IEEE TRANSACTIONS ON COMPUTERS, Vol. 69, No. 2, 8859334, 01.02.2020, p. 185-197.

Research output: Contribution to journalArticle

Harvard

Singh, AK, Basireddy, KR, Prakash, A, Merrett, GV & Al-Hashimi, BM 2020, 'Collaborative Adaptation for Energy-Efficient Heterogeneous Mobile SoCs', IEEE TRANSACTIONS ON COMPUTERS, vol. 69, no. 2, 8859334, pp. 185-197. https://doi.org/10.1109/TC.2019.2943855

APA

Singh, A. K., Basireddy, K. R., Prakash, A., Merrett, G. V., & Al-Hashimi, B. M. (2020). Collaborative Adaptation for Energy-Efficient Heterogeneous Mobile SoCs. IEEE TRANSACTIONS ON COMPUTERS, 69(2), 185-197. [8859334]. https://doi.org/10.1109/TC.2019.2943855

Vancouver

Singh AK, Basireddy KR, Prakash A, Merrett GV, Al-Hashimi BM. Collaborative Adaptation for Energy-Efficient Heterogeneous Mobile SoCs. IEEE TRANSACTIONS ON COMPUTERS. 2020 Feb 1;69(2):185-197. 8859334. https://doi.org/10.1109/TC.2019.2943855

Author

Singh, A.K. ; Basireddy, K.R. ; Prakash, A. ; Merrett, G.V. ; Al-Hashimi, Bashir M. / Collaborative Adaptation for Energy-Efficient Heterogeneous Mobile SoCs. In: IEEE TRANSACTIONS ON COMPUTERS. 2020 ; Vol. 69, No. 2. pp. 185-197.

Bibtex Download

@article{1d794e42278a49a2a10b618a0386b63d,
title = "Collaborative Adaptation for Energy-Efficient Heterogeneous Mobile SoCs",
abstract = "Heterogeneous Mobile System-on-Chips (SoCs) containing CPU and GPU cores are becoming prevalent in embedded computing, and they need to execute applications concurrently. However, existing run-time management approaches do not perform adaptive mapping and thread-partitioning of applications while exploiting both CPU and GPU cores at the same time. In this paper, we propose an adaptive mapping and thread-partitioning approach for energy-efficient execution of concurrent OpenCL applications on both CPU and GPU cores while satisfying performance requirements. To start execution of concurrent applications, the approach makes mapping (number of cores and operating frequencies) and partitioning (distribution of threads between CPU and GPU) decisions to satisfy performance requirements for each application. The mapping and partitioning decisions are made by having a collaboration between the CPU and GPU cores' processing capabilities such that balanced execution can be performed. During execution, adaptation is triggered when new application(s) arrive, or an executing one finishes, that frees cores. The adaptation process identifies a new mapping and thread-partitioning in a similar collaborative manner for remaining applications provided it leads to an improvement in energy efficiency. The proposed approach is experimentally validated on the Odroid-XU3 hardware platform with varying set of applications. Results show an average energy saving of 37%, compared to existing approaches while satisfying the performance requirements.",
keywords = "SoC, adaptation, concurrent execution, energy-efficiency, heterogeneous computing",
author = "A.K. Singh and K.R. Basireddy and A. Prakash and G.V. Merrett and Al-Hashimi, {Bashir M.}",
note = "cited By 0",
year = "2020",
month = feb,
day = "1",
doi = "10.1109/TC.2019.2943855",
language = "English",
volume = "69",
pages = "185--197",
journal = "IEEE TRANSACTIONS ON COMPUTERS",
issn = "0018-9340",
publisher = "IEEE Computer Society",
number = "2",

}

RIS (suitable for import to EndNote) Download

TY - JOUR

T1 - Collaborative Adaptation for Energy-Efficient Heterogeneous Mobile SoCs

AU - Singh, A.K.

AU - Basireddy, K.R.

AU - Prakash, A.

AU - Merrett, G.V.

AU - Al-Hashimi, Bashir M.

N1 - cited By 0

PY - 2020/2/1

Y1 - 2020/2/1

N2 - Heterogeneous Mobile System-on-Chips (SoCs) containing CPU and GPU cores are becoming prevalent in embedded computing, and they need to execute applications concurrently. However, existing run-time management approaches do not perform adaptive mapping and thread-partitioning of applications while exploiting both CPU and GPU cores at the same time. In this paper, we propose an adaptive mapping and thread-partitioning approach for energy-efficient execution of concurrent OpenCL applications on both CPU and GPU cores while satisfying performance requirements. To start execution of concurrent applications, the approach makes mapping (number of cores and operating frequencies) and partitioning (distribution of threads between CPU and GPU) decisions to satisfy performance requirements for each application. The mapping and partitioning decisions are made by having a collaboration between the CPU and GPU cores' processing capabilities such that balanced execution can be performed. During execution, adaptation is triggered when new application(s) arrive, or an executing one finishes, that frees cores. The adaptation process identifies a new mapping and thread-partitioning in a similar collaborative manner for remaining applications provided it leads to an improvement in energy efficiency. The proposed approach is experimentally validated on the Odroid-XU3 hardware platform with varying set of applications. Results show an average energy saving of 37%, compared to existing approaches while satisfying the performance requirements.

AB - Heterogeneous Mobile System-on-Chips (SoCs) containing CPU and GPU cores are becoming prevalent in embedded computing, and they need to execute applications concurrently. However, existing run-time management approaches do not perform adaptive mapping and thread-partitioning of applications while exploiting both CPU and GPU cores at the same time. In this paper, we propose an adaptive mapping and thread-partitioning approach for energy-efficient execution of concurrent OpenCL applications on both CPU and GPU cores while satisfying performance requirements. To start execution of concurrent applications, the approach makes mapping (number of cores and operating frequencies) and partitioning (distribution of threads between CPU and GPU) decisions to satisfy performance requirements for each application. The mapping and partitioning decisions are made by having a collaboration between the CPU and GPU cores' processing capabilities such that balanced execution can be performed. During execution, adaptation is triggered when new application(s) arrive, or an executing one finishes, that frees cores. The adaptation process identifies a new mapping and thread-partitioning in a similar collaborative manner for remaining applications provided it leads to an improvement in energy efficiency. The proposed approach is experimentally validated on the Odroid-XU3 hardware platform with varying set of applications. Results show an average energy saving of 37%, compared to existing approaches while satisfying the performance requirements.

KW - SoC

KW - adaptation

KW - concurrent execution

KW - energy-efficiency

KW - heterogeneous computing

UR - http://www.scopus.com/inward/record.url?scp=85078346190&partnerID=8YFLogxK

U2 - 10.1109/TC.2019.2943855

DO - 10.1109/TC.2019.2943855

M3 - Article

VL - 69

SP - 185

EP - 197

JO - IEEE TRANSACTIONS ON COMPUTERS

JF - IEEE TRANSACTIONS ON COMPUTERS

SN - 0018-9340

IS - 2

M1 - 8859334

ER -

View graph of relations

© 2020 King's College London | Strand | London WC2R 2LS | England | United Kingdom | Tel +44 (0)20 7836 5454