Towards a Predictive Energy Model for HPC Runtime Systems Using Supervised Learning

Gence Ozer; Sarthak Garg; Neda Davoudi; Gabrielle Poerwawinata; Matthias Maiterth; Alessio Netti; Daniele Tafani

doi:10.1007/978-3-030-48340-1_48

Towards a Predictive Energy Model for HPC Runtime Systems Using Supervised Learning

Gence Ozer, Sarthak Garg, Neda Davoudi, Gabrielle Poerwawinata, Matthias Maiterth, Alessio Netti, Daniele Tafani

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

6 Scopus citations

Abstract

High-Performance Computing systems collect vast amounts of operational data with the employment of monitoring frameworks, often augmented with additional information from schedulers and runtime systems. This amount of data can be used and turned into a benefit for operational requirements, rather than being a data pool for post-mortem analysis. This work focuses on deriving a model with supervised learning which enables optimal selection of CPU frequency during the execution of a job, with the objective of minimizing the energy consumption of a HPC system. Our model is trained utilizing sensor data and performance metrics collected with two distinct open-source frameworks for monitoring and runtime optimization. Our results show good prediction of CPU power draw and number of instructions retired on realistic dynamic runtime settings within a relatively low error margin.

Original language	English
Title of host publication	Euro-Par 2019
Subtitle of host publication	Parallel Processing Workshops - International Workshops, Revised Selected Papers
Editors	Ulrich Schwardmann, Christian Boehme, Dora B. Heras, Valeria Cardellini, Emmanuel Jeannot, Antonio Salis, Claudio Schifanella, Ravi Reddy Manumachu, Dieter Schwamborn, Laura Ricci, Oh Sangyoon, Thomas Gruber, Laura Antonelli, Stephen L. Scott
Publisher	Springer
Pages	626-638
Number of pages	13
ISBN (Print)	9783030483395
DOIs	https://doi.org/10.1007/978-3-030-48340-1_48
State	Published - 2020
Externally published	Yes
Event	25th International European Conference on Parallel and Distributed Computing, EuroPar 2019 - Göttingen, Germany Duration: Aug 26 2019 → Aug 30 2019

Publication series

Name	Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume	11997 LNCS
ISSN (Print)	0302-9743
ISSN (Electronic)	1611-3349

Conference

Conference	25th International European Conference on Parallel and Distributed Computing, EuroPar 2019
Country/Territory	Germany
City	Göttingen
Period	08/26/19 → 08/30/19

Keywords

DVFS
Energy efficiency
Monitoring systems
Random forest
Runtime systems

Access to Document

10.1007/978-3-030-48340-1_48

Cite this

Ozer, G., Garg, S., Davoudi, N., Poerwawinata, G., Maiterth, M., Netti, A., & Tafani, D. (2020). Towards a Predictive Energy Model for HPC Runtime Systems Using Supervised Learning. In U. Schwardmann, C. Boehme, D. B. Heras, V. Cardellini, E. Jeannot, A. Salis, C. Schifanella, R. R. Manumachu, D. Schwamborn, L. Ricci, O. Sangyoon, T. Gruber, L. Antonelli, & S. L. Scott (Eds.), Euro-Par 2019: Parallel Processing Workshops - International Workshops, Revised Selected Papers (pp. 626-638). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 11997 LNCS). Springer. https://doi.org/10.1007/978-3-030-48340-1_48

Ozer, Gence ; Garg, Sarthak ; Davoudi, Neda et al. / Towards a Predictive Energy Model for HPC Runtime Systems Using Supervised Learning. Euro-Par 2019: Parallel Processing Workshops - International Workshops, Revised Selected Papers. editor / Ulrich Schwardmann ; Christian Boehme ; Dora B. Heras ; Valeria Cardellini ; Emmanuel Jeannot ; Antonio Salis ; Claudio Schifanella ; Ravi Reddy Manumachu ; Dieter Schwamborn ; Laura Ricci ; Oh Sangyoon ; Thomas Gruber ; Laura Antonelli ; Stephen L. Scott. Springer, 2020. pp. 626-638 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).

@inproceedings{316d7d125ed94545b5e846a75cda75b9,

title = "Towards a Predictive Energy Model for HPC Runtime Systems Using Supervised Learning",

abstract = "High-Performance Computing systems collect vast amounts of operational data with the employment of monitoring frameworks, often augmented with additional information from schedulers and runtime systems. This amount of data can be used and turned into a benefit for operational requirements, rather than being a data pool for post-mortem analysis. This work focuses on deriving a model with supervised learning which enables optimal selection of CPU frequency during the execution of a job, with the objective of minimizing the energy consumption of a HPC system. Our model is trained utilizing sensor data and performance metrics collected with two distinct open-source frameworks for monitoring and runtime optimization. Our results show good prediction of CPU power draw and number of instructions retired on realistic dynamic runtime settings within a relatively low error margin.",

keywords = "DVFS, Energy efficiency, Monitoring systems, Random forest, Runtime systems",

author = "Gence Ozer and Sarthak Garg and Neda Davoudi and Gabrielle Poerwawinata and Matthias Maiterth and Alessio Netti and Daniele Tafani",

note = "Publisher Copyright: {\textcopyright} 2020, Springer Nature Switzerland AG.; 25th International European Conference on Parallel and Distributed Computing, EuroPar 2019 ; Conference date: 26-08-2019 Through 30-08-2019",

year = "2020",

doi = "10.1007/978-3-030-48340-1_48",

language = "English",

isbn = "9783030483395",

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

publisher = "Springer",

pages = "626--638",

editor = "Ulrich Schwardmann and Christian Boehme and {B. Heras}, Dora and Valeria Cardellini and Emmanuel Jeannot and Antonio Salis and Claudio Schifanella and Manumachu, {Ravi Reddy} and Dieter Schwamborn and Laura Ricci and Oh Sangyoon and Thomas Gruber and Laura Antonelli and Scott, {Stephen L.}",

booktitle = "Euro-Par 2019",

}

Ozer, G, Garg, S, Davoudi, N, Poerwawinata, G, Maiterth, M, Netti, A & Tafani, D 2020, Towards a Predictive Energy Model for HPC Runtime Systems Using Supervised Learning. in U Schwardmann, C Boehme, D B. Heras, V Cardellini, E Jeannot, A Salis, C Schifanella, RR Manumachu, D Schwamborn, L Ricci, O Sangyoon, T Gruber, L Antonelli & SL Scott (eds), Euro-Par 2019: Parallel Processing Workshops - International Workshops, Revised Selected Papers. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 11997 LNCS, Springer, pp. 626-638, 25th International European Conference on Parallel and Distributed Computing, EuroPar 2019, Göttingen, Germany, 08/26/19. https://doi.org/10.1007/978-3-030-48340-1_48

Towards a Predictive Energy Model for HPC Runtime Systems Using Supervised Learning. / Ozer, Gence; Garg, Sarthak; Davoudi, Neda et al.
Euro-Par 2019: Parallel Processing Workshops - International Workshops, Revised Selected Papers. ed. / Ulrich Schwardmann; Christian Boehme; Dora B. Heras; Valeria Cardellini; Emmanuel Jeannot; Antonio Salis; Claudio Schifanella; Ravi Reddy Manumachu; Dieter Schwamborn; Laura Ricci; Oh Sangyoon; Thomas Gruber; Laura Antonelli; Stephen L. Scott. Springer, 2020. p. 626-638 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 11997 LNCS).

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

TY - GEN

T1 - Towards a Predictive Energy Model for HPC Runtime Systems Using Supervised Learning

AU - Ozer, Gence

AU - Garg, Sarthak

AU - Davoudi, Neda

AU - Poerwawinata, Gabrielle

AU - Maiterth, Matthias

AU - Netti, Alessio

AU - Tafani, Daniele

PY - 2020

Y1 - 2020

N2 - High-Performance Computing systems collect vast amounts of operational data with the employment of monitoring frameworks, often augmented with additional information from schedulers and runtime systems. This amount of data can be used and turned into a benefit for operational requirements, rather than being a data pool for post-mortem analysis. This work focuses on deriving a model with supervised learning which enables optimal selection of CPU frequency during the execution of a job, with the objective of minimizing the energy consumption of a HPC system. Our model is trained utilizing sensor data and performance metrics collected with two distinct open-source frameworks for monitoring and runtime optimization. Our results show good prediction of CPU power draw and number of instructions retired on realistic dynamic runtime settings within a relatively low error margin.

AB - High-Performance Computing systems collect vast amounts of operational data with the employment of monitoring frameworks, often augmented with additional information from schedulers and runtime systems. This amount of data can be used and turned into a benefit for operational requirements, rather than being a data pool for post-mortem analysis. This work focuses on deriving a model with supervised learning which enables optimal selection of CPU frequency during the execution of a job, with the objective of minimizing the energy consumption of a HPC system. Our model is trained utilizing sensor data and performance metrics collected with two distinct open-source frameworks for monitoring and runtime optimization. Our results show good prediction of CPU power draw and number of instructions retired on realistic dynamic runtime settings within a relatively low error margin.

KW - DVFS

KW - Energy efficiency

KW - Monitoring systems

KW - Random forest

KW - Runtime systems

UR - http://www.scopus.com/inward/record.url?scp=85086229363&partnerID=8YFLogxK

U2 - 10.1007/978-3-030-48340-1_48

DO - 10.1007/978-3-030-48340-1_48

M3 - Conference contribution

AN - SCOPUS:85086229363

SN - 9783030483395

T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

SP - 626

EP - 638

BT - Euro-Par 2019

A2 - Schwardmann, Ulrich

A2 - Boehme, Christian

A2 - B. Heras, Dora

A2 - Cardellini, Valeria

A2 - Jeannot, Emmanuel

A2 - Salis, Antonio

A2 - Schifanella, Claudio

A2 - Manumachu, Ravi Reddy

A2 - Schwamborn, Dieter

A2 - Ricci, Laura

A2 - Sangyoon, Oh

A2 - Gruber, Thomas

A2 - Antonelli, Laura

A2 - Scott, Stephen L.

PB - Springer

T2 - 25th International European Conference on Parallel and Distributed Computing, EuroPar 2019

Y2 - 26 August 2019 through 30 August 2019

ER -

Ozer G, Garg S, Davoudi N, Poerwawinata G, Maiterth M, Netti A et al. Towards a Predictive Energy Model for HPC Runtime Systems Using Supervised Learning. In Schwardmann U, Boehme C, B. Heras D, Cardellini V, Jeannot E, Salis A, Schifanella C, Manumachu RR, Schwamborn D, Ricci L, Sangyoon O, Gruber T, Antonelli L, Scott SL, editors, Euro-Par 2019: Parallel Processing Workshops - International Workshops, Revised Selected Papers. Springer. 2020. p. 626-638. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)). doi: 10.1007/978-3-030-48340-1_48

Towards a Predictive Energy Model for HPC Runtime Systems Using Supervised Learning

Abstract

Publication series

Conference

Keywords

Access to Document

Other files and links

Fingerprint

Cite this