high performance computing on graphics processing units: hgpu.org

hgpu.org » Mixed precision

Large Scale Language Modeling: Converging on 40GB of Text in Four Hours

Raul Puri, Robert Kirby, Nikolai Yakovenko, Bryan Catanzaro

View

Download (PDF)

Source codes

Tags: Computer science, CUDA, Deep learning, LSTM, Mixed precision, Neural networks, nVidia, Package, Tesla V100, Torch

August 11, 2018 by hgpu

Aspect-Driven Mixed-Precision Tuning Targeting GPUs

Ricardo Nobre, Luis Reis, Joao Bispo, Tiago Carvalho, Joao M.P. Cardoso, Stefano Cherubin, Giovanni Agosta

View

Download (PDF)

Source codes

Tags: AMD Vega 10 XT, ATI, Code generation, Computer science, Mixed precision, nVidia, OpenCL, Package

June 13, 2018 by hgpu

NVIDIA Tensor Core Programmability, Performance & Precision

Stefano Markidis, Steven Wei Der Chien, Erwin Laure, Ivy Bo Peng, Jeffrey S. Vetter

View

Download (PDF)

Tags: Computer science, CUDA, Mixed precision, nVidia, Performance, Tesla V100

March 17, 2018 by hgpu

Investigating Half Precision Arithmetic to Accelerate Dense Linear System Solvers

Azzam Haidar, Panruo Wu, Stanimire Tomov, Jack Dongarra

View

Download (PDF)

Tags: Algorithms, Artificial intelligence, BLAS, Computer science, Linar Algebra, Mixed precision, Neural networks, nVidia, Tesla P100

December 10, 2017 by hgpu

Mixed Precision Solver Scalable to 16000 MPI Processes for Lattice Quantum Chromodynamics Simulations on the Oakforest-PACS System

Taisuke Boku, Ishikawa Ken-Ichi, Yoshinobu Kuramashi, Lawrence Meadows

View

Download (PDF)

Source codes

Tags: Computational Physics, High Energy Physics - Lattice, Intel Xeon Phi, Mixed precision, MPI, OpenMP, Package, Physics, QCD

September 28, 2017 by hgpu

Automatic code generation methods applied to numerical linear algebra in high performance computing

M. Ian Masliah

View

Download (PDF)

Tags: Algorithms, Code generation, Computer science, CUDA, Heterogeneous systems, Linear Algebra, Mixed precision, nVidia, nVidia Tegra TX1, Programming techniques, Tesla C2075, Tesla K40, Thesis

November 16, 2016 by hgpu

Efficient and portable acceleration of quantum chemical many-body methods in mixed floating point precision using OpenACC compiler directives

Janus Juul Eriksen

View

Download (PDF)

Tags: Chemical Physics, Chemistry, Mixed precision, nVidia, OpenACC, Physics, Tesla K40

September 27, 2016 by hgpu

Stability and Performance of Various Singular Value QR Implementations on Multicore CPU with a GPU

Ichitaro Yamazaki, Stanimire Tomov, Jack Dongarra

View

Download (PDF)

Tags: BLAS, Computer science, CUDA, Linear Algebra, Matrix decomposition, Mixed precision, Numerical Analysis, nVidia, Performance, Tesla K20

July 16, 2016 by hgpu

A mixed precision semi-Lagrangian algorithm and its performance on accelerators

Lukas Einkemmer

View

Download (PDF)

Tags: Computational Physics, Computer science, CUDA, Intel Xeon Phi, Mathematical Software, Mixed precision, nVidia, OpenMP, Tesla K80

March 25, 2016 by hgpu

Optimising Reconfigurable Systems for Real-time Applications

Thomas Chun Pong Chau

View

Download (PDF)

Source codes

Tags: Algorithms, Computer science, CUDA, FPGA, Machine learning, Mixed precision, nVidia, Package, Tesla C2070, Thesis

August 7, 2015 by hgpu

Towards an automatic generation of dense linear algebra solvers on parallel architectures

Marc Baboulin, Joel Falcou, Ian Masliah

View

Download (PDF)

Tags: Computer science, CUDA, Linear Algebra, Mixed precision, nVidia, Tesla C2075

December 20, 2014 by hgpu

Mixed-precision orthogonalization scheme and its case studies with CA-GMRES on a GPU

Ichitaro Yamazaki, Stanimire Tomov, Tingxing Dong, Jack Dongarra

View

Download (PDF)

Tags: Algorithms, Computer science, CUDA, Mixed precision, nVidia, Tesla K20, Tesla K40, Tesla M2090

July 29, 2014 by hgpu

Characterising Across-Stack Optimisations for Deep Convolutional Neural Networks

Using the Tsetlin Machine to Learn Human-Interpretable Rules for High-Accuracy Text Categorization with Medical Applications

ZUCL: A ZYNQ UltraScale+ Framework for OpenCL HLS Applications

CL-kmeans: Efficient k-Means in OpenCL

Efficient and Scalable k-Means on GPUs

Doctor AI: Interpretable Deep Learning for Modeling Electronic Health Records

pyJac: a Python package that generates source code used to analytically calculate chemical kinetics Jacobian matrices

Using SIMD and SIMT vectorization to evaluate sparse chemical kinetic Jacobian matrices and thermochemical source terms

CUTLASS: CUDA Templates for Linear Algebra Subroutines

Implementing Strassen’s Algorithm with CUTLASS on NVIDIA Volta GPUs

Integer sorting on multicore

A study of integer sorting on multicores

Kernel Tuner: A search-optimizing GPU code auto-tuner

libxsmm: Library targeting Intel Architecture for specialized dense and sparse matrix operations, and deep learning primitives

Anatomy Of High-Performance Deep Learning Convolutions On SIMD Architectures

See all packages

* * *

2019

February

25-28

Shanghai, China

International Conference on Image, Video and Signal Processing (IVSP), 2019

2018

November

20-22

Khalifa University, Abu Dhabi, UAE

International Joint Conference on Signals, Systems and Computers (CSSC), 2018

2018

December

13-15

Wellington, New Zealand

International Joint Conference on Robotics, Automation and Mechatronics (JCRAM), 2018

2018

December

10-12

Wellington, New Zealand

International Joint Conference on Robotics and Artificial Intelligence (JCRAI), 2018

2018

December

13-15

Wellington, New Zealand

International Joint Conference on Computer Vision and Pattern Recognition (CCVPR), 2018

Add event

See all events

high performance computing on graphics processing units: hgpu.org

Large Scale Language Modeling: Converging on 40GB of Text in Four Hours

Aspect-Driven Mixed-Precision Tuning Targeting GPUs

NVIDIA Tensor Core Programmability, Performance & Precision

Investigating Half Precision Arithmetic to Accelerate Dense Linear System Solvers

Mixed Precision Solver Scalable to 16000 MPI Processes for Lattice Quantum Chromodynamics Simulations on the Oakforest-PACS System

Automatic code generation methods applied to numerical linear algebra in high performance computing

Efficient and portable acceleration of quantum chemical many-body methods in mixed floating point precision using OpenACC compiler directives

Stability and Performance of Various Singular Value QR Implementations on Multicore CPU with a GPU

A mixed precision semi-Lagrangian algorithm and its performance on accelerators

Optimising Reconfigurable Systems for Real-time Applications

Towards an automatic generation of dense linear algebra solvers on parallel architectures

Mixed-precision orthogonalization scheme and its case studies with CA-GMRES on a GPU

Recent source codes

Characterising Across-Stack Optimisations for Deep Convolutional Neural Networks

Using the Tsetlin Machine to Learn Human-Interpretable Rules for High-Accuracy Text Categorization with Medical Applications

ZUCL: A ZYNQ UltraScale+ Framework for OpenCL HLS Applications

CL-kmeans: Efficient k-Means in OpenCL

Doctor AI: Interpretable Deep Learning for Modeling Electronic Health Records

pyJac: a Python package that generates source code used to analytically calculate chemical kinetics Jacobian matrices

CUTLASS: CUDA Templates for Linear Algebra Subroutines

Integer sorting on multicore

Kernel Tuner: A search-optimizing GPU code auto-tuner

libxsmm: Library targeting Intel Architecture for specialized dense and sparse matrix operations, and deep learning primitives

Most viewed papers (last 30 days)

Featured events

International Conference on Image, Video and Signal Processing (IVSP), 2019

International Joint Conference on Signals, Systems and Computers (CSSC), 2018

International Joint Conference on Robotics, Automation and Mechatronics (JCRAM), 2018

International Joint Conference on Robotics and Artificial Intelligence (JCRAI), 2018

International Joint Conference on Computer Vision and Pattern Recognition (CCVPR), 2018