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

Research output: Journal contributionsJournal articlesResearchpeer-review

Authors

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.

Original languageEnglish
Article numbere6877
JournalConcurrency and Computation: Practice and Experience
Volume35
Issue number17
Number of pages16
ISSN1532-0626
DOIs
Publication statusPublished - 01.08.2023
Externally publishedYes

Bibliographical note

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.

    Research areas

  • dark-silicon, design space exploration, GPUs, heterogeneous computing, performance predictors
  • Business informatics

DOI

Recently viewed

Activities

  1. Winding the Sociotechnical Mangle of Organization
  2. BDSM Sagacity: embodying complexity
  3. Preliminary selection of experimental techniques in Subtask D
  4. "Curious and Concerned" – A mixed-methods study of teacher educators’ AI literacy, usage experience, and perceptions
  5. Conference of SIG4 'Higher Education' & SIG17 'Qualitative and Quantitative Approaches to Learning and Instruction'
  6. Efficacy of an Internet-based problem-solving training for teachers: Results of a randomized controlled trial.
  7. Coauthoring an interorganizational collaboration: Exploring multi-voicedness and introducing spatiotemporal orientations
  8. Assessing cognitive abilities for L2 learning: A review of accessible tools
  9. Project Workshop on "Worker Flows, Match Quality, and Productivity" - 2019
  10. Plenary Speaker at the 2022 5th IEEE International Conference on Information Communication and Signal Processing (ICICSP 2022). Plenary entitled "Soft Optimal Computing Techniques to Identify and Control Surface Roughness in Manufacturing".
  11. A Tool for Applications: Wavelet Packets
  12. E-learning module on “Participation” in the context of IWRM – “Social Science” Part
  13. Workshop „Meta-Image Day 2012”
  14. The 22nd International Semantic Web Conference (ISWC 2023)
  15. Monitoring Phenological phases of the crops using TerraSAR-X Spotlight Images based on Factor Analysis
  16. Unit Root & Cointegration Testing Conference 2005
  17. Robots versus Machines
  18. Explaining the learning progress in mathematics of retained students and low-achieving students

Publications

  1. Evaluating a Bayesian Student Model of Decimal Misconceptions
  2. The identification of up-And downstream industries using input-output tables and a firm-level application to minority shareholdings
  3. The structure of emotions in learning situations
  4. Introduction to the Design, Implementation, and Management of Digital Government Policies and Strategies Minitrack
  5. Detection of coherent oceanic structures via transfer operators
  6. Introduction: The representative turn in EU studies
  7. Taking notes as a strategy for solving reality-based tasks in mathematics
  8. Outperformed by a Computer? - Comparing Human Decisions to Reinforcement Learning Agents, Assigning Lot Sizes in a Learning Factory
  9. Closed-form Solution for the Direct Kinematics Problem of the Planar 3-RPR Parallel Mechanism
  10. Towards productive functions?
  11. Factored MDPs for detecting topics of user sessions
  12. Investigating the Effect of Noise Elimination on LSTM Models for Financial Markets Prediction Using Kalman Filter and Wavelet Transform
  13. Who can receive the pass? A computational model for quantifying availability in soccer
  14. An Outcome-Oriented, Social-Ecological Framework for Assessing Protected Area Effectiveness
  15. Combining Evaluative and Generative Diagnosis in ActiveMath
  16. “Ideation is Fine, but Execution is Key”
  17. Towards a spatial understanding of identity play
  18. Introduction
  19. Foreign bias in institutional portfolio allocation
  20. Active and semi-supervised data domain description
  21. Double-fading support - A training approach to complex software systems
  22. Trajectory tracking using MPC and a velocity observer for flat actuator systems in automotive applications
  23. ℓp-norm multiple kernel learning
  24. Individual Scans Fusion in Virtual Knowledge Base for Navigation of Mobile Robotic Group with 3D TVS
  25. Functional Richness and Relative Resilience of Bird Communities in Regions with Different Land Use Intensities
  26. Pressure fault recognition and compensation with an adaptive feedforward regulator in a controlled hybrid actuator within engine applications