Software-defined PMC for Runtime Power Management of a Many-core Neuromorphic PlatformCitation formats

  • External authors:
  • Indar Sugiarto
  • Delong Shang
  • Amit Kumar Singh
  • Bassem Ouni
  • Geoff Merrett
  • Bashir Al-Hashimi

Standard

Software-defined PMC for Runtime Power Management of a Many-core Neuromorphic Platform. / Sugiarto, Indar; Shang, Delong; Singh, Amit Kumar; Ouni, Bassem; Merrett, Geoff; Al-Hashimi, Bashir; Furber, Stephen.

2017. 1-6 Paper presented at 12th IEEE International Conference on Computer Engineering and Systems, Cairo, Egypt.

Research output: Contribution to conferencePaperpeer-review

Harvard

Sugiarto, I, Shang, D, Singh, AK, Ouni, B, Merrett, G, Al-Hashimi, B & Furber, S 2017, 'Software-defined PMC for Runtime Power Management of a Many-core Neuromorphic Platform', Paper presented at 12th IEEE International Conference on Computer Engineering and Systems, Cairo, Egypt, 19/12/17 - 20/12/17 pp. 1-6.

APA

Sugiarto, I., Shang, D., Singh, A. K., Ouni, B., Merrett, G., Al-Hashimi, B., & Furber, S. (Accepted/In press). Software-defined PMC for Runtime Power Management of a Many-core Neuromorphic Platform. 1-6. Paper presented at 12th IEEE International Conference on Computer Engineering and Systems, Cairo, Egypt.

Vancouver

Sugiarto I, Shang D, Singh AK, Ouni B, Merrett G, Al-Hashimi B et al. Software-defined PMC for Runtime Power Management of a Many-core Neuromorphic Platform. 2017. Paper presented at 12th IEEE International Conference on Computer Engineering and Systems, Cairo, Egypt.

Author

Sugiarto, Indar ; Shang, Delong ; Singh, Amit Kumar ; Ouni, Bassem ; Merrett, Geoff ; Al-Hashimi, Bashir ; Furber, Stephen. / Software-defined PMC for Runtime Power Management of a Many-core Neuromorphic Platform. Paper presented at 12th IEEE International Conference on Computer Engineering and Systems, Cairo, Egypt.6 p.

Bibtex

@conference{b5ec96482047405b8a101abf1c36c169,
title = "Software-defined PMC for Runtime Power Management of a Many-core Neuromorphic Platform",
abstract = "This paper presents an approach to provide a Run-time Management (RTM) system for a many-core neuromorphic platform. RTM frameworks are commonly used to achieve an energy saving while satisfying application performance requirements. In commodity processors, the RTM can be implemented by utilizing the output of Performance Monitoring Counters (PMCs) to control the frequency of the processor's clock. However, many neuromorphic platforms such as SpiNNaker do not have PMC units; thus, we propose a software-defined PMC that can be implemented using standard programming tool-chains in such platforms. In this paper, we evaluate several control strategies for RTM in SpiNNaker. These control programs are equivalent with governors in standard operating systems such as Linux. For evaluation, we use the RTM with several image processing applications. The results show that our proposed method, called Improved-Conservative, produces the lowest thermal risk and energy consumption while achieving the same performance as other adaptive governors.",
keywords = "PMC, RTM, Many-core, Neuromorphic, SpiNNaker",
author = "Indar Sugiarto and Delong Shang and Singh, {Amit Kumar} and Bassem Ouni and Geoff Merrett and Bashir Al-Hashimi and Stephen Furber",
year = "2017",
language = "English",
pages = "1--6",
note = "12th IEEE International Conference on Computer Engineering and Systems, ICCES 2017 ; Conference date: 19-12-2017 Through 20-12-2017",
url = "http://www.icces.org.eg/",

}

RIS

TY - CONF

T1 - Software-defined PMC for Runtime Power Management of a Many-core Neuromorphic Platform

AU - Sugiarto, Indar

AU - Shang, Delong

AU - Singh, Amit Kumar

AU - Ouni, Bassem

AU - Merrett, Geoff

AU - Al-Hashimi, Bashir

AU - Furber, Stephen

N1 - Conference code: 42632

PY - 2017

Y1 - 2017

N2 - This paper presents an approach to provide a Run-time Management (RTM) system for a many-core neuromorphic platform. RTM frameworks are commonly used to achieve an energy saving while satisfying application performance requirements. In commodity processors, the RTM can be implemented by utilizing the output of Performance Monitoring Counters (PMCs) to control the frequency of the processor's clock. However, many neuromorphic platforms such as SpiNNaker do not have PMC units; thus, we propose a software-defined PMC that can be implemented using standard programming tool-chains in such platforms. In this paper, we evaluate several control strategies for RTM in SpiNNaker. These control programs are equivalent with governors in standard operating systems such as Linux. For evaluation, we use the RTM with several image processing applications. The results show that our proposed method, called Improved-Conservative, produces the lowest thermal risk and energy consumption while achieving the same performance as other adaptive governors.

AB - This paper presents an approach to provide a Run-time Management (RTM) system for a many-core neuromorphic platform. RTM frameworks are commonly used to achieve an energy saving while satisfying application performance requirements. In commodity processors, the RTM can be implemented by utilizing the output of Performance Monitoring Counters (PMCs) to control the frequency of the processor's clock. However, many neuromorphic platforms such as SpiNNaker do not have PMC units; thus, we propose a software-defined PMC that can be implemented using standard programming tool-chains in such platforms. In this paper, we evaluate several control strategies for RTM in SpiNNaker. These control programs are equivalent with governors in standard operating systems such as Linux. For evaluation, we use the RTM with several image processing applications. The results show that our proposed method, called Improved-Conservative, produces the lowest thermal risk and energy consumption while achieving the same performance as other adaptive governors.

KW - PMC

KW - RTM

KW - Many-core

KW - Neuromorphic

KW - SpiNNaker

M3 - Paper

SP - 1

EP - 6

T2 - 12th IEEE International Conference on Computer Engineering and Systems

Y2 - 19 December 2017 through 20 December 2017

ER -