Dynamic workload peak detection for slack management

A. Milutinovic; Kees Goossens; Gerardus Johannes Maria Smit

doi:10.1109/SOCC.2009.5335679

Dynamic workload peak detection for slack management

A. Milutinovic
, Kees Goossens
, Gerardus Johannes Maria Smit

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › Academic › peer-review

234 Downloads (Pure)

Abstract

In this paper an analytical study on dynamism and possibilities on slack exploitation by dynamic power management is presented. We introduce a specific workload decomposition method for work required for (streaming) application processing data tokens (e.g. video frames) with work behaviour patterns as a mix of periodic and aperiodic patterns. It offers efficient and computationally light method for speculation on considerable work variations and its exploitation in energy saving techniques. It is used by a dynamic power management policy which has low overhead and reduces both requirements for buffering space, and deadline misses (increase QoS). We evaluate our policy in experiments on MPEG4 decoding of several different input sequences and present results.

Original language	Undefined
Title of host publication	Proceedings of International Symposium on System-on-Chip, 2009. SOC 2009.
Editors	Jan Kuper, J. Kuper
Place of Publication	Tampere
Publisher	IEEE
Pages	42-47
Number of pages	6
ISBN (Print)	978-1-4244-4465-6
DOIs	https://doi.org/10.1109/SOCC.2009.5335679
Publication status	Published - 5 Oct 2009
Event	Proceedings of International Symposium on System-on-Chip, 2009. SOC 2009. - Tampere, Finland Duration: 5 Oct 2009 → 7 Oct 2009

Publication series

Name
Publisher	IEEE Computer Society

Conference

Conference	Proceedings of International Symposium on System-on-Chip, 2009. SOC 2009.
Period	5/10/09 → 7/10/09
Other	5-7 Oct 2009

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

SDG 7 Affordable and Clean Energy

Keywords

METIS-264158
IR-68663
EWI-16533
CAES-EEA: Efficient Embedded Architectures
Low power
Embedded Systems
Power Management

Access to Document

10.1109/SOCC.2009.5335679

33
33

Cite this

@inproceedings{2ca41ff159f549cb909ac1ea4cd7f0a2,

title = "Dynamic workload peak detection for slack management",

abstract = "In this paper an analytical study on dynamism and possibilities on slack exploitation by dynamic power management is presented. We introduce a specific workload decomposition method for work required for (streaming) application processing data tokens (e.g. video frames) with work behaviour patterns as a mix of periodic and aperiodic patterns. It offers efficient and computationally light method for speculation on considerable work variations and its exploitation in energy saving techniques. It is used by a dynamic power management policy which has low overhead and reduces both requirements for buffering space, and deadline misses (increase QoS). We evaluate our policy in experiments on MPEG4 decoding of several different input sequences and present results.",

keywords = "METIS-264158, IR-68663, EWI-16533, CAES-EEA: Efficient Embedded Architectures, Low power, Embedded Systems, Power Management",

author = "A. Milutinovic and Kees Goossens and Smit, \{Gerardus Johannes Maria\}",

note = "10.1109/SOCC.2009.5335679 ; Proceedings of International Symposium on System-on-Chip, 2009. SOC 2009. ; Conference date: 05-10-2009 Through 07-10-2009",

year = "2009",

month = oct,

day = "5",

doi = "10.1109/SOCC.2009.5335679",

language = "Undefined",

isbn = "978-1-4244-4465-6",

publisher = "IEEE",

pages = "42--47",

editor = "Jan Kuper and J. Kuper",

booktitle = "Proceedings of International Symposium on System-on-Chip, 2009. SOC 2009.",

address = "United States",

}

Milutinovic, A, Goossens, K & Smit, GJM 2009, Dynamic workload peak detection for slack management. in J Kuper & J Kuper (eds), Proceedings of International Symposium on System-on-Chip, 2009. SOC 2009.., 10.1109/SOCC.2009.5335679, IEEE, Tampere, pp. 42-47, Proceedings of International Symposium on System-on-Chip, 2009. SOC 2009., 5/10/09. https://doi.org/10.1109/SOCC.2009.5335679

Dynamic workload peak detection for slack management. / Milutinovic, A.; Goossens, Kees; Smit, Gerardus Johannes Maria.
Proceedings of International Symposium on System-on-Chip, 2009. SOC 2009.. ed. / Jan Kuper; J. Kuper. Tampere: IEEE, 2009. p. 42-47 10.1109/SOCC.2009.5335679.

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › Academic › peer-review

TY - GEN

T1 - Dynamic workload peak detection for slack management

AU - Milutinovic, A.

AU - Goossens, Kees

AU - Smit, Gerardus Johannes Maria

N1 - 10.1109/SOCC.2009.5335679

PY - 2009/10/5

Y1 - 2009/10/5

N2 - In this paper an analytical study on dynamism and possibilities on slack exploitation by dynamic power management is presented. We introduce a specific workload decomposition method for work required for (streaming) application processing data tokens (e.g. video frames) with work behaviour patterns as a mix of periodic and aperiodic patterns. It offers efficient and computationally light method for speculation on considerable work variations and its exploitation in energy saving techniques. It is used by a dynamic power management policy which has low overhead and reduces both requirements for buffering space, and deadline misses (increase QoS). We evaluate our policy in experiments on MPEG4 decoding of several different input sequences and present results.

AB - In this paper an analytical study on dynamism and possibilities on slack exploitation by dynamic power management is presented. We introduce a specific workload decomposition method for work required for (streaming) application processing data tokens (e.g. video frames) with work behaviour patterns as a mix of periodic and aperiodic patterns. It offers efficient and computationally light method for speculation on considerable work variations and its exploitation in energy saving techniques. It is used by a dynamic power management policy which has low overhead and reduces both requirements for buffering space, and deadline misses (increase QoS). We evaluate our policy in experiments on MPEG4 decoding of several different input sequences and present results.

KW - METIS-264158

KW - IR-68663

KW - EWI-16533

KW - CAES-EEA: Efficient Embedded Architectures

KW - Low power

KW - Embedded Systems

KW - Power Management

U2 - 10.1109/SOCC.2009.5335679

DO - 10.1109/SOCC.2009.5335679

M3 - Conference contribution

SN - 978-1-4244-4465-6

SP - 42

EP - 47

BT - Proceedings of International Symposium on System-on-Chip, 2009. SOC 2009.

A2 - Kuper, Jan

A2 - Kuper, J.

PB - IEEE

CY - Tampere

T2 - Proceedings of International Symposium on System-on-Chip, 2009. SOC 2009.

Y2 - 5 October 2009 through 7 October 2009

ER -