Distributed power management of real-time applications on a GALS multiprocessor SOC

Onderzoeksoutput: Hoofdstuk in Boek/Rapport/CongresprocedureConferentiebijdrageAcademicpeer review

1 Citaat (Scopus)

Uittreksel

It is generally desirable to reduce the power consumption of embedded systems. Dynamic Voltage and Frequency Scaling (DVFS) is a commonly applied technique to achieve power reduction at the cost of computational performance. Multiprocessor System on Chips (MPSoCs) can have multiple voltage and frequency domains, e.g. per-core. When DVFS is applied to real-time applications, the effects must be accounted for in the associated formal timing model. In this work, we contribute our distributed multi-core run-time power-management technique for real-time dataflow applications that uses per-core lookup-tables to select low-power DVFS operating points that meet the application's timing requirement. We describe in detail how timing slack is observed locally at run-time on each core and is used to select a local DVFS operating point that meets the application's timing requirement. We further describe our static off-line formal analysis technique to generate these per-core lookup-tables that link timing slack to low-power DVFS operating points. We provide an experimental analysis of our proposed technique using an H.263 decoder application that is mapped onto an FPGA prototyped hardware platform.
TaalEngels
Titel2015 International Conference on Embedded Software (EMSOFT), 4-9 October 2015, Amsterdam, The Netherlands
Plaats van productiePiscataway
UitgeverijInstitute of Electrical and Electronics Engineers
Pagina's147-156
DOI's
StatusGepubliceerd - okt 2015

Vingerafdruk

Table lookup
Embedded systems
Field programmable gate arrays (FPGA)
Electric power utilization
Dynamic frequency scaling
Voltage scaling
Power management
Hardware
Electric potential
System-on-chip

Citeer dit

Nelson, A., & Goossens, K. (2015). Distributed power management of real-time applications on a GALS multiprocessor SOC. In 2015 International Conference on Embedded Software (EMSOFT), 4-9 October 2015, Amsterdam, The Netherlands (blz. 147-156). Piscataway: Institute of Electrical and Electronics Engineers. DOI: 10.1109/EMSOFT.2015.7318270
Nelson, Andrew ; Goossens, Kees. / Distributed power management of real-time applications on a GALS multiprocessor SOC. 2015 International Conference on Embedded Software (EMSOFT), 4-9 October 2015, Amsterdam, The Netherlands. Piscataway : Institute of Electrical and Electronics Engineers, 2015. blz. 147-156
@inproceedings{d4c83b7e4de844c49c6d45f022c6c546,
title = "Distributed power management of real-time applications on a GALS multiprocessor SOC",
abstract = "It is generally desirable to reduce the power consumption of embedded systems. Dynamic Voltage and Frequency Scaling (DVFS) is a commonly applied technique to achieve power reduction at the cost of computational performance. Multiprocessor System on Chips (MPSoCs) can have multiple voltage and frequency domains, e.g. per-core. When DVFS is applied to real-time applications, the effects must be accounted for in the associated formal timing model. In this work, we contribute our distributed multi-core run-time power-management technique for real-time dataflow applications that uses per-core lookup-tables to select low-power DVFS operating points that meet the application's timing requirement. We describe in detail how timing slack is observed locally at run-time on each core and is used to select a local DVFS operating point that meets the application's timing requirement. We further describe our static off-line formal analysis technique to generate these per-core lookup-tables that link timing slack to low-power DVFS operating points. We provide an experimental analysis of our proposed technique using an H.263 decoder application that is mapped onto an FPGA prototyped hardware platform.",
author = "Andrew Nelson and Kees Goossens",
year = "2015",
month = "10",
doi = "10.1109/EMSOFT.2015.7318270",
language = "English",
pages = "147--156",
booktitle = "2015 International Conference on Embedded Software (EMSOFT), 4-9 October 2015, Amsterdam, The Netherlands",
publisher = "Institute of Electrical and Electronics Engineers",
address = "United States",

}

Nelson, A & Goossens, K 2015, Distributed power management of real-time applications on a GALS multiprocessor SOC. in 2015 International Conference on Embedded Software (EMSOFT), 4-9 October 2015, Amsterdam, The Netherlands. Institute of Electrical and Electronics Engineers, Piscataway, blz. 147-156. DOI: 10.1109/EMSOFT.2015.7318270

Distributed power management of real-time applications on a GALS multiprocessor SOC. / Nelson, Andrew; Goossens, Kees.

2015 International Conference on Embedded Software (EMSOFT), 4-9 October 2015, Amsterdam, The Netherlands. Piscataway : Institute of Electrical and Electronics Engineers, 2015. blz. 147-156.

Onderzoeksoutput: Hoofdstuk in Boek/Rapport/CongresprocedureConferentiebijdrageAcademicpeer review

TY - GEN

T1 - Distributed power management of real-time applications on a GALS multiprocessor SOC

AU - Nelson,Andrew

AU - Goossens,Kees

PY - 2015/10

Y1 - 2015/10

N2 - It is generally desirable to reduce the power consumption of embedded systems. Dynamic Voltage and Frequency Scaling (DVFS) is a commonly applied technique to achieve power reduction at the cost of computational performance. Multiprocessor System on Chips (MPSoCs) can have multiple voltage and frequency domains, e.g. per-core. When DVFS is applied to real-time applications, the effects must be accounted for in the associated formal timing model. In this work, we contribute our distributed multi-core run-time power-management technique for real-time dataflow applications that uses per-core lookup-tables to select low-power DVFS operating points that meet the application's timing requirement. We describe in detail how timing slack is observed locally at run-time on each core and is used to select a local DVFS operating point that meets the application's timing requirement. We further describe our static off-line formal analysis technique to generate these per-core lookup-tables that link timing slack to low-power DVFS operating points. We provide an experimental analysis of our proposed technique using an H.263 decoder application that is mapped onto an FPGA prototyped hardware platform.

AB - It is generally desirable to reduce the power consumption of embedded systems. Dynamic Voltage and Frequency Scaling (DVFS) is a commonly applied technique to achieve power reduction at the cost of computational performance. Multiprocessor System on Chips (MPSoCs) can have multiple voltage and frequency domains, e.g. per-core. When DVFS is applied to real-time applications, the effects must be accounted for in the associated formal timing model. In this work, we contribute our distributed multi-core run-time power-management technique for real-time dataflow applications that uses per-core lookup-tables to select low-power DVFS operating points that meet the application's timing requirement. We describe in detail how timing slack is observed locally at run-time on each core and is used to select a local DVFS operating point that meets the application's timing requirement. We further describe our static off-line formal analysis technique to generate these per-core lookup-tables that link timing slack to low-power DVFS operating points. We provide an experimental analysis of our proposed technique using an H.263 decoder application that is mapped onto an FPGA prototyped hardware platform.

U2 - 10.1109/EMSOFT.2015.7318270

DO - 10.1109/EMSOFT.2015.7318270

M3 - Conference contribution

SP - 147

EP - 156

BT - 2015 International Conference on Embedded Software (EMSOFT), 4-9 October 2015, Amsterdam, The Netherlands

PB - Institute of Electrical and Electronics Engineers

CY - Piscataway

ER -

Nelson A, Goossens K. Distributed power management of real-time applications on a GALS multiprocessor SOC. In 2015 International Conference on Embedded Software (EMSOFT), 4-9 October 2015, Amsterdam, The Netherlands. Piscataway: Institute of Electrical and Electronics Engineers. 2015. blz. 147-156. Beschikbaar vanaf, DOI: 10.1109/EMSOFT.2015.7318270