The energy/frequency convexity rule: Modeling and experimental validation on mobile devices

Karel De Vogeleer; Gerard Memmi; Pierre Jouvelot; Fabien Coelho

doi:10.1007/978-3-642-55224-3_74

The energy/frequency convexity rule: Modeling and experimental validation on mobile devices

Karel De Vogeleer
, Gerard Memmi
, Pierre Jouvelot
, Fabien Coelho

CNRS LTCI
Mines ParisTech

Research output: Contribution to journal › Conference article › peer-review

62 Citations (Scopus)

Abstract

This paper provides both theoretical and experimental evidence for the existence of an Energy/Frequency Convexity Rule, which relates energy consumption and CPU frequency on mobile devices. We monitored a typical smartphone running a specific computing-intensive kernel of multiple nested loops written in C using a high-resolution power gauge. Data gathered during a week-long acquisition campaign suggest that energy consumed per input element is strongly correlated with CPU frequency, and, more interestingly, the curve exhibits a clear minimum over a 0.2 GHz to 1.6 GHz window. We provide and motivate an analytical model for this behavior, which fits well with the data. Our work should be of clear interest to researchers focusing on energy usage and minimization for mobile devices, and provide new insights for optimization opportunities.

Original language	English
Pages (from-to)	793-803
Number of pages	11
Journal	Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume	8384 LNCS
Issue number	PART 1
DOIs	https://doi.org/10.1007/978-3-642-55224-3_74
Publication status	Published - 1 Jan 2014
Externally published	Yes
Event	10th International Conference on Parallel Processing and Applied Mathematics, PPAM 2013 - Warsaw, Poland Duration: 8 Sept 2013 → 11 Sept 2013

UN SDGs

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

SDG 7 Affordable and Clean Energy

Keywords

Bit-reverse algorithm
DVFS
Energy consumption and modeling
Execution time modeling
Power consumption
Smartphone

Access to Document

10.1007/978-3-642-55224-3_74

Cite this

@article{c50baf1d5d4947d49198f45549ff58f3,

title = "The energy/frequency convexity rule: Modeling and experimental validation on mobile devices",

abstract = "This paper provides both theoretical and experimental evidence for the existence of an Energy/Frequency Convexity Rule, which relates energy consumption and CPU frequency on mobile devices. We monitored a typical smartphone running a specific computing-intensive kernel of multiple nested loops written in C using a high-resolution power gauge. Data gathered during a week-long acquisition campaign suggest that energy consumed per input element is strongly correlated with CPU frequency, and, more interestingly, the curve exhibits a clear minimum over a 0.2 GHz to 1.6 GHz window. We provide and motivate an analytical model for this behavior, which fits well with the data. Our work should be of clear interest to researchers focusing on energy usage and minimization for mobile devices, and provide new insights for optimization opportunities.",

keywords = "Bit-reverse algorithm, DVFS, Energy consumption and modeling, Execution time modeling, Power consumption, Smartphone",

author = "\{De Vogeleer\}, Karel and Gerard Memmi and Pierre Jouvelot and Fabien Coelho",

year = "2014",

month = jan,

day = "1",

doi = "10.1007/978-3-642-55224-3\_74",

language = "English",

volume = "8384 LNCS",

pages = "793--803",

journal = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",

issn = "0302-9743",

publisher = "Springer Science and Business Media Deutschland GmbH",

number = "PART 1",

note = "10th International Conference on Parallel Processing and Applied Mathematics, PPAM 2013 ; Conference date: 08-09-2013 Through 11-09-2013",

}

De Vogeleer, K, Memmi, G, Jouvelot, P & Coelho, F 2014, 'The energy/frequency convexity rule: Modeling and experimental validation on mobile devices', Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 8384 LNCS, no. PART 1, pp. 793-803. https://doi.org/10.1007/978-3-642-55224-3_74

The energy/frequency convexity rule: Modeling and experimental validation on mobile devices. / De Vogeleer, Karel; Memmi, Gerard; Jouvelot, Pierre et al.
In: Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), Vol. 8384 LNCS, No. PART 1, 01.01.2014, p. 793-803.

Research output: Contribution to journal › Conference article › peer-review

TY - JOUR

T1 - The energy/frequency convexity rule

T2 - 10th International Conference on Parallel Processing and Applied Mathematics, PPAM 2013

AU - De Vogeleer, Karel

AU - Memmi, Gerard

AU - Jouvelot, Pierre

AU - Coelho, Fabien

PY - 2014/1/1

Y1 - 2014/1/1

N2 - This paper provides both theoretical and experimental evidence for the existence of an Energy/Frequency Convexity Rule, which relates energy consumption and CPU frequency on mobile devices. We monitored a typical smartphone running a specific computing-intensive kernel of multiple nested loops written in C using a high-resolution power gauge. Data gathered during a week-long acquisition campaign suggest that energy consumed per input element is strongly correlated with CPU frequency, and, more interestingly, the curve exhibits a clear minimum over a 0.2 GHz to 1.6 GHz window. We provide and motivate an analytical model for this behavior, which fits well with the data. Our work should be of clear interest to researchers focusing on energy usage and minimization for mobile devices, and provide new insights for optimization opportunities.

AB - This paper provides both theoretical and experimental evidence for the existence of an Energy/Frequency Convexity Rule, which relates energy consumption and CPU frequency on mobile devices. We monitored a typical smartphone running a specific computing-intensive kernel of multiple nested loops written in C using a high-resolution power gauge. Data gathered during a week-long acquisition campaign suggest that energy consumed per input element is strongly correlated with CPU frequency, and, more interestingly, the curve exhibits a clear minimum over a 0.2 GHz to 1.6 GHz window. We provide and motivate an analytical model for this behavior, which fits well with the data. Our work should be of clear interest to researchers focusing on energy usage and minimization for mobile devices, and provide new insights for optimization opportunities.

KW - Bit-reverse algorithm

KW - DVFS

KW - Energy consumption and modeling

KW - Execution time modeling

KW - Power consumption

KW - Smartphone

U2 - 10.1007/978-3-642-55224-3_74

DO - 10.1007/978-3-642-55224-3_74

M3 - Conference article

AN - SCOPUS:84901249973

SN - 0302-9743

VL - 8384 LNCS

SP - 793

EP - 803

JO - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

JF - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

IS - PART 1

Y2 - 8 September 2013 through 11 September 2013

ER -

The energy/frequency convexity rule: Modeling and experimental validation on mobile devices

Abstract

UN SDGs

Keywords

Access to Document

Fingerprint

Cite this