Matrix Product (GEMM) Performance Data from GPUs

Joubert, Wayne; Palmer, Eric

doi:10.13139/OLCF/1819195

Title: Matrix Product (GEMM) Performance Data from GPUs

Dataset
Other Related Research

Abstract

Timing data for mixed precision GEMM matrix product operations on several GPU models, including NVIDIA V100 and A100, AMD MI100 and Intel P580. Also data from machine learning model training on this data using Scikit-learn.

Authors:: Joubert, Wayne; Palmer, Eric

Publication Date:: 2021-09-09

DOE Contract Number:: DE-AC05-00OR22725

Research Org.:: Oak Ridge National Lab. (ORNL), Oak Ridge, TN (United States). Oak Ridge Leadership Computing Facility (OLCF)

Sponsoring Org.:: USDOE Office of Science (SC); Office of Science (SC), Advanced Scientific Computing Research (ASCR) (SC-21)

Subject:: 97 MATHEMATICS AND COMPUTING

Keywords:: GEMM matrix product operations; computer system benchmarking; GPU Graphical Processing Units

OSTI Identifier:: 1819195

DOI:: https://doi.org/10.13139/OLCF/1819195

Citation Formats


                    Joubert, Wayne, and Palmer, Eric. Matrix Product (GEMM) Performance Data from GPUs.  United States: N. p., 2021. 
        Web.  doi:10.13139/OLCF/1819195.

Copy to clipboard


                    Joubert, Wayne, & Palmer, Eric. Matrix Product (GEMM) Performance Data from GPUs.  United States.  doi:https://doi.org/10.13139/OLCF/1819195

Copy to clipboard


                    Joubert, Wayne, and Palmer, Eric. 2021.  
"Matrix Product (GEMM) Performance Data from GPUs".  United States.  doi:https://doi.org/10.13139/OLCF/1819195.  https://www.osti.gov/servlets/purl/1819195. Pub date:Thu Sep 09 00:00:00 EDT 2021

Copy to clipboard


                    
@article{osti_1819195,

  title        = {Matrix Product (GEMM) Performance Data from GPUs},

  author       = {Joubert, Wayne and Palmer, Eric},

  abstractNote = {Timing data for mixed precision GEMM matrix product operations on several GPU models, including NVIDIA V100 and A100, AMD MI100 and Intel P580. Also data from machine learning model training on this data using Scikit-learn.},

  doi          = {10.13139/OLCF/1819195},

  journal      = {},

  number       = ,

  volume       = ,

  place        = {United States},

  year         = {2021},

  month        = {9}

}

Copy to clipboard

Dataset:

View Dataset

DOI: https://doi.org/10.13139/OLCF/1819195

Save / Share:

Export Metadata

Save to My Library

Similar records in DOE Data Explorer and OSTI.GOV collections:

Performance Analysis of Memory Transfers and GEMM Subroutines on NVIDIA Tesla GPU Cluster

Conference Allada, Veerendra, Benjegerdes, Troy ; Bode, Brett

Commodity clusters augmented with application accelerators are evolving as competitive high performance computing systems. The Graphical Processing Unit (GPU) with a very high arithmetic density and performance per price ratio is a good platform for the scientific application acceleration. In addition to the interconnect bottlenecks among the cluster compute nodes, the cost of memory copies between the host and the GPU device have to be carefully amortized to improve the overall efficiency of the application. Scientific applications also rely on efficient implementation of the BAsic Linear Algebra Subroutines (BLAS), among which the General Matrix Multiply (GEMM) is considered as themore »« less
Data from: Can the results of biodiversity-ecosystem productivity studies be translated to bioenergy production?

Dataset Dickson, Timothy L. ; Gross, Katherine L.

Biodiversity experiments show that increases in plant diversity can lead to greater biomass production, and some researchers suggest that high diversity plantings should be used for bioenergy production. However, many methods used in past biodiversity experiments are impractical for bioenergy plantings. For example, biodiversity experiments often use intensive management such as hand weeding to maintain low diversity plantings and exclude unplanted species, but this would not be done for bioenergy plantings. Also, biodiversity experiments generally use high seeding densities that would be too expensive for bioenergy plantings. Here we report the effects of biodiversity on biomass production from two studiesmore »« less
Performance Data from a 1-Meter Cross-flow Turbine with High Deflection Hydrofoils

Dataset Marone, Nicole ; Wosnik, Martin ; Barrington, Matthew ; ...

Performance data of a 1-meter diameter cross-flow tidal turbine consisting of three NACA 0018 blades with two support struts with high deflection hydrofoils. Data was collected at the University of New Hampshire Jere A. Chase Ocean Engineering Lab within the tow tank. Three turbine parameters were varied: the blade materials, blade shape, and support strut position. A detailed description of the testing set-up and data files contained within the compressed HDF.zip file is in the 'ReadMe.txt' file.
Data from: "Snowmelt Timing Regulates Community Composition, Phenology, and Physiological Performance of Alpine Plants"

Dataset Winkler, Daniel E. ; Butz, Ramona J. ; Germino, Matthew J. ; ...

This archive contains data that were used to support conclusions drawn in “Snowmelt Timing Regulates Community Composition, Phenology, and Physiological Performance of Alpine Plants”, by Winkler et al., 2018. Data were collected throughout the 2009 growing season on Niwot Ridge, Colorado, before the site became part of the Alpine Treeline Warming Experiment (ATWE). Geospatial files are included in this archive to provide additional locational context. The files in this data package consist of five comma-separated-values (.csv) files, one keyhole markup language (.kml) file, and two ESRI shapefiles (.shp). The .csv files can be opened by Microsoft Excel, R, or anymore »« less
Search for Higgs boson pair production in the $b \bar{b} b \bar{b}$ final state from $p p$ collisions at $\sqrt{s} = 8$ TeV with the ATLAS detector

Dataset Aad, Georges ; The ATLAS collaboration

Similar Records