high performance computing on graphics processing units: hgpu.org

hgpu.org » CUBLAS

Implementing Strassen’s Algorithm with CUTLASS on NVIDIA Volta GPUs

Jianyu Huang, Chenhan D. Yu, Robert A. van de Geijn

View

Download (PDF)

Source codes

Tags: Algorithms, Computer science, CUBLAS, CUDA, Matrix multiplication, nVidia, Package, Tesla V100

September 2, 2018 by hgpu

An In-depth Performance Characterization of CPU- and GPU-based DNN Training on Modern Architectures

Ammar Ahmad Awan, Hari Subramoni, Dhabaleswar K. Panda

View

Download (PDF)

Tags: Benchmarking, Caffe, Computer science, CUBLAS, CUDA, Deep learning, Intel Xeon Phi, Machine learning, nVidia, Tela K40, Tesla K80, Tesla P100

December 24, 2017 by hgpu

Out-of-core Implementation for Accelerator Kernels on Heterogeneous Clouds

Hamidreza Khaleghzadeh, Ziming Zhong, Ravi Reddy, Alexey Lastovetsky

View

Download (PDF)

Source codes

Tags: BLAS, Cloud, Computer science, CUBLAS, CUDA, FPGA, Heterogeneous systems, Intel Xeon Phi, Matrix multiplication, nVidia, OpenCL, Package, Virtualization

September 16, 2017 by hgpu

Acceleration of Linear Finite-Difference Poisson-Boltzmann Methods on Graphics Processing Units

Ruxi Qi, Wesley M. Botello-Smith, Ray Luo

View

Download (PDF)

Tags: Biomolecules, Biophysics, Boltzmann equation, Computational Physics, CUBLAS, CUDA, Electrostatics, Molecular simulation, nVidia, nVidia GeForce GTX 980 Ti, Physics, Poisson Boltzmann

April 11, 2017 by hgpu

Improving the Performance of Fully Connected Neural Networks by Out-of-Place Matrix Transpose

Shaohuai Shi, Pengfei Xu, Xiaowen Chu

View

Download (PDF)

Source codes

Tags: Algorithms, BLAS, Caffe, Computer science, CUBLAS, CUDA, Deep learning, Linear Algebra, Matrix multiplication, Neural networks, nVidia, nVidia GeForce GTX 1080, Package, Performance

February 14, 2017 by hgpu

GPU performance analysis of a nodal discontinuous Galerkin method for acoustic and elastic models

Axel Modave, Amik St-Cyr, Tim Warburton

View

Download (PDF)

Tags: BLAS, Computational Physics, CUBLAS, CUDA, FEM, Finite element method, Geoscience, Linear Algebra, nVidia, nVidia GeForce GTX 980, OCCA, Physics, Profiling, Seismic modeling, Seismology

March 1, 2016 by hgpu

BLASX: A High Performance Level-3 BLAS Library for Heterogeneous Multi-GPU Computing

Linnan Wang, Wei Wu, Jianxiong Xiao, Yi Yang

View

Download (PDF)

Source codes

Tags: Benchmarking, Computer science, CUBLAS, CUDA, Heterogeneous systems, Linear Algebra, nVidia, nVidia GeForce GTX Titan X, Package, Performance, Tesla K40

October 22, 2015 by hgpu

linalg: Matrix Computations in Apache Spark

Reza Bosagh Zadeh, Xiangrui Meng, Burak Yavuz, Aaron Staple, Li Pu, Shivaram Venkataraman, Evan Sparks, Alexander Ulanov, Matei Zaharia

View

Download (PDF)

Source codes

Tags: Algorithms, Benchmarking, Computer science, CUBLAS, CUDA, Linear Algebra, Machine learning, Matrix multiplication, nVidia, Package, Scala, Tesla M2050

September 15, 2015 by hgpu

Optimizing strassen matrix multiply on GPUs

Ayaz ul Hasan Khan, Mayez Al-Mouhamed, Allam Fatayer

View

Download (PDF)

Tags: Algorithms, Computer science, CUBLAS, CUDA, Data parallelism, Matrix multiplication, nVidia, Tesla K20

August 11, 2015 by hgpu

Framework for Batched and GPU-resident Factorization Algorithms Applied to Block Householder Transformations

Azzam Haidar, Tingxing "Tim" Dong, Stanimire Tomov, Piotr Luszczek, Jack Dongarra

View

Download (PDF)

Tags: Algorithms, Computer science, CUBLAS, CUDA, Factorization, nVidia, Tesla K40

April 12, 2015 by hgpu

Batched Matrix Computations on Hardware Accelerators

Azzam Haidar, Tingxing Dong, Piotr Luszczek, Stanimire Tomov, Jack Dongarra

View

Download (PDF)

Tags: Algorithms, Computer science, CUBLAS, CUDA, Factorization, Linear Algebra, nVidia, Tesla K40

April 12, 2015 by hgpu

Data driven scheduling approach for the multi-node multi-GPU Cholesky decomposition

Yuki Tsujita, Toshio Endo

View

Download (PDF)

Tags: Computer science, CUBLAS, CUDA, Heterogeneous systems, Linear Algebra, nVidia, Task scheduling, Tesla K20

March 25, 2015 by hgpu

Over-synchronization in GPU Programs

profile_util: simple C++ library to report memory, timing, thread affinity

Profile Util library: A quick and easy way to get MPI, OpenMP and GPU runtime information

MaC: Mamba CPU performance predictor

MambaCPU: Enhanced Correlation Mining with State Space Models for CPU Performance Prediction

Mixed-precision finite element kernels

Mixed-precision finite element kernels and assembly: Rounding error analysis and hardware acceleration

SHMT: The Simultaneous and Heterogenous Multithreading project

General-Purpose Computing on Tensor Processors

Superpipeline: A Universal Approach for Reducing GPU Memory Usage in Large Models

EnergyUCB-Bandit

Online Energy Optimization in GPUs: A Multi-Armed Bandit Approach

See all packages

* * *

high performance computing on graphics processing units: hgpu.org

Implementing Strassen’s Algorithm with CUTLASS on NVIDIA Volta GPUs

An In-depth Performance Characterization of CPU- and GPU-based DNN Training on Modern Architectures

Acceleration of Linear Finite-Difference Poisson-Boltzmann Methods on Graphics Processing Units

Improving the Performance of Fully Connected Neural Networks by Out-of-Place Matrix Transpose

GPU performance analysis of a nodal discontinuous Galerkin method for acoustic and elastic models

BLASX: A High Performance Level-3 BLAS Library for Heterogeneous Multi-GPU Computing

linalg: Matrix Computations in Apache Spark

Optimizing strassen matrix multiply on GPUs

Framework for Batched and GPU-resident Factorization Algorithms Applied to Block Householder Transformations

Batched Matrix Computations on Hardware Accelerators

Data driven scheduling approach for the multi-node multi-GPU Cholesky decomposition

Recent source codes

Kokkidio: Kokkos+Eigen - write fast, readable multi-platform code

LLaMA-Inference-Bench

MATAR

Over-synchronization in GPU Programs

profile_util: simple C++ library to report memory, timing, thread affinity

MaC: Mamba CPU performance predictor

Mixed-precision finite element kernels

SHMT: The Simultaneous and Heterogenous Multithreading project

Superpipeline: A Universal Approach for Reducing GPU Memory Usage in Large Models

EnergyUCB-Bandit

Most viewed papers (last 30 days)