Performance predictors for graphics processing units applied to dark-silicon-aware design space exploration

Research output: Journal contributionsJournal articlesResearchpeer-review

Standard

Performance predictors for graphics processing units applied to dark-silicon-aware design space exploration. / Sonohata, Rhayssa; Arigoni, Danillo Christi A.; Fernandes, Eraldo Rezende et al.
In: Concurrency and Computation: Practice and Experience, Vol. 35, No. 17, e6877, 01.08.2023.

Research output: Journal contributionsJournal articlesResearchpeer-review

Harvard

APA

Vancouver

Sonohata R, Arigoni DCA, Fernandes ER, Ribeiro dos Santos R, Dessandre Duenha L. Performance predictors for graphics processing units applied to dark-silicon-aware design space exploration. Concurrency and Computation: Practice and Experience. 2023 Aug 1;35(17):e6877. Epub 2022 Mar 4. doi: 10.1002/cpe.6877

Bibtex

@article{d26e304306bd4cd2b493275aecbed0fa,
title = "Performance predictors for graphics processing units applied to dark-silicon-aware design space exploration",
abstract = "The limitations on the scalability of computer systems imposed by the dark-silicon effects are so severe that they support the extensive use of heterogeneity such as the GP-GPU for general purpose processing. Performance simulators of GP-GPU heterogeneous systems aim to provide performance accuracy at the cost of execution time. In this work, we handle time-consuming simulations of design space exploration systems based on GPUs. We have developed performance predictors based on machine learning (ML) algorithms and evaluated them in accuracy and throughput (number of predictions per second). We measure model accuracy through the mean absolute percentage error (MAPE) and the model efficiency through a throughput metric (millions of predictions per second). Our experiments revealed that decision trees predictors are the most promising regarding accuracy and efficiency. We applied the best predictors into the MultiExplorer, a dark silicon-aware design space exploration tool that allows designers to explore the architecture and microarchitecture of multicore/manycore system design.",
keywords = "dark-silicon, design space exploration, GPUs, heterogeneous computing, performance predictors, Business informatics",
author = "Rhayssa Sonohata and Arigoni, {Danillo Christi A.} and Fernandes, {Eraldo Rezende} and {Ribeiro dos Santos}, Ricardo and {Dessandre Duenha}, Liana",
note = "Special Issue: WSCAD 2020. PDCAT 2020/PDCAT‐PAAP 2020 This study was financed in part by FUNDECT and Coordena{\c c}{\~a}o de Aperfei{\c c}oamento de Pessoal de N{\'i}vel Superior ‐ Brasil (CAPES) ‐ Finance Code 001. ",
year = "2023",
month = aug,
day = "1",
doi = "10.1002/cpe.6877",
language = "English",
volume = "35",
journal = "Concurrency and Computation: Practice and Experience",
issn = "1532-0626",
publisher = "John Wiley & Sons Ltd.",
number = "17",

}

RIS

TY - JOUR

T1 - Performance predictors for graphics processing units applied to dark-silicon-aware design space exploration

AU - Sonohata, Rhayssa

AU - Arigoni, Danillo Christi A.

AU - Fernandes, Eraldo Rezende

AU - Ribeiro dos Santos, Ricardo

AU - Dessandre Duenha, Liana

N1 - Special Issue: WSCAD 2020. PDCAT 2020/PDCAT‐PAAP 2020 This study was financed in part by FUNDECT and Coordenação de Aperfeiçoamento de Pessoal de Nível Superior ‐ Brasil (CAPES) ‐ Finance Code 001.

PY - 2023/8/1

Y1 - 2023/8/1

N2 - The limitations on the scalability of computer systems imposed by the dark-silicon effects are so severe that they support the extensive use of heterogeneity such as the GP-GPU for general purpose processing. Performance simulators of GP-GPU heterogeneous systems aim to provide performance accuracy at the cost of execution time. In this work, we handle time-consuming simulations of design space exploration systems based on GPUs. We have developed performance predictors based on machine learning (ML) algorithms and evaluated them in accuracy and throughput (number of predictions per second). We measure model accuracy through the mean absolute percentage error (MAPE) and the model efficiency through a throughput metric (millions of predictions per second). Our experiments revealed that decision trees predictors are the most promising regarding accuracy and efficiency. We applied the best predictors into the MultiExplorer, a dark silicon-aware design space exploration tool that allows designers to explore the architecture and microarchitecture of multicore/manycore system design.

AB - The limitations on the scalability of computer systems imposed by the dark-silicon effects are so severe that they support the extensive use of heterogeneity such as the GP-GPU for general purpose processing. Performance simulators of GP-GPU heterogeneous systems aim to provide performance accuracy at the cost of execution time. In this work, we handle time-consuming simulations of design space exploration systems based on GPUs. We have developed performance predictors based on machine learning (ML) algorithms and evaluated them in accuracy and throughput (number of predictions per second). We measure model accuracy through the mean absolute percentage error (MAPE) and the model efficiency through a throughput metric (millions of predictions per second). Our experiments revealed that decision trees predictors are the most promising regarding accuracy and efficiency. We applied the best predictors into the MultiExplorer, a dark silicon-aware design space exploration tool that allows designers to explore the architecture and microarchitecture of multicore/manycore system design.

KW - dark-silicon

KW - design space exploration

KW - GPUs

KW - heterogeneous computing

KW - performance predictors

KW - Business informatics

UR - http://www.scopus.com/inward/record.url?scp=85125564653&partnerID=8YFLogxK

U2 - 10.1002/cpe.6877

DO - 10.1002/cpe.6877

M3 - Journal articles

AN - SCOPUS:85125564653

VL - 35

JO - Concurrency and Computation: Practice and Experience

JF - Concurrency and Computation: Practice and Experience

SN - 1532-0626

IS - 17

M1 - e6877

ER -

DOI

Recently viewed

Researchers

  1. Georg Reischauer

Publications

  1. Chapter 9: Particular Remedies for Non-performance: Section 2: Withholding Performance
  2. Challenges for postdocs in Germany and beyond:
  3. Principals between exploitation and exploration
  4. Conceptions of problem solving mathematics teaching
  5. Effects Of Different Order Processing Strategies On Operating Curves Of Logistic Models
  6. Communicating CCS
  7. Integrating regional perceptions into climate change adaptation
  8. Analysis of the construction of an autonomous robot to improve its energy efficiency when traveling through irregular terrain
  9. Chapter 9: Particular Remedies for Non-performance: Section 3: Termination of Contract
  10. Bridging scenario planning and backcasting
  11. Release of monomers from four different composite materials after halogen and LED curing
  12. Recurring patterns and blueprints of industrial symbioses as structural units for an it tool
  13. The Integration of Wheelchair Users in Team Handball
  14. Framework, Drivers and Information Needs for Creating Business Cases for Sustainability
  15. Taking stock–Three years of addressing societal challenges on community level through action research
  16. A Smart Sensing Architecture for Misalignment Measurements
  17. Nutrients addition regulates temperature sensitivity of maize straw mineralization
  18. The Rhythm Method
  19. Portfolio optimisation of power futures market
  20. Consumer Preferences for Local Food: Testing an Extended Norm Taxonomy
  21. Prekäre Existenz
  22. Dealing with availability and response expectations: Are older employees at an advantage and why?
  23. Die Todesmärsche von Buchenwald