GPGPU performance estimation for frequency scaling using cross-benchmarking

Qiang WANG, Chengjian Liu, Xiaowen CHU

Research output: Chapter in book/report/conference proceedingConference contributionpeer-review

1 Citation (Scopus)

Abstract

Dynamic Voltage and Frequency Scaling (DVFS) on General-Purpose Graphics Processing Units (GPGPUs) is now becoming one of the most significant techniques to balance computational performance and energy consumption. However, there are still few fast and accurate models for predicting GPU kernel execution time under different core and memory frequency settings, which is important to determine the best frequency configuration for energy saving. Accordingly, a novel GPGPU performance estimation model with both core and memory frequency scaling is herein proposed. We design a cross-benchmarking suite, which simulates kernels with a wide range of instruction distributions. The synthetic kernels generated by this suite can be used for model pre-training or as supplementary training samples. Then we apply two different machine learning algorithms, Support Vector Regression (SVR) and Gradient Boosting Decision Tree (GBDT), to study the correlation between kernel performance counters and kernel performance. The models trained only with our cross-benchmarking suite achieve satisfying accuracy (16%∼22% mean absolute error) on 24 unseen real application kernels. Validated on three modern GPUs with a wide frequency scaling range, by using a collection of 24 real application kernels, the proposed model is able to achieve accurate results (5.1%, 2.8%, 6.5% mean absolute error) for the target GPUs (GTX 980, Titan X Pascal and Tesla P100).

Original languageEnglish
Title of host publicationGPGPU 2020 - Proceedings of the 2020 General Purpose Processing Using GPU
PublisherAssociation for Computing Machinery, Inc
Pages31-40
Number of pages10
ISBN (Electronic)9781450370257
DOIs
Publication statusPublished - 23 Feb 2020
Event13th Annual Workshop on General Purpose Processing using Graphics Processing Unit, GPGPU 2020 - San Diego, United States
Duration: 23 Feb 202023 Feb 2020

Publication series

NameGPGPU 2020 - Proceedings of the 2020 General Purpose Processing Using GPU

Conference

Conference13th Annual Workshop on General Purpose Processing using Graphics Processing Unit, GPGPU 2020
Country/TerritoryUnited States
CitySan Diego
Period23/02/2023/02/20

Scopus Subject Areas

  • Computer Graphics and Computer-Aided Design
  • Hardware and Architecture
  • Software

User-Defined Keywords

  • Dynamic Voltage and Frequency Scaling
  • GPU Performance Modeling
  • Graphics Processing Units
  • Machine Learning

Fingerprint

Dive into the research topics of 'GPGPU performance estimation for frequency scaling using cross-benchmarking'. Together they form a unique fingerprint.

Cite this