high performance computing on graphics processing units: hgpu.org

Jehandad Khan, Paul Fultz, Artem Tamazov, Daniel Lowell, Chao Liu, Michael Melesse, Murali Nandhimandalam, Kamil Nasyrov, Ilya Perminov, Tejash Shah, Vasilii Filippov, Jing Zhang, Jing Zhou, Bragadeesh Natarajan, Mayank Daga

View

Download (PDF)

Source codes

Tags: Computer science, Deep learning, Neural networks, OpenCL, Package

October 6, 2019 by hgpu

Exascale Deep Learning for Scientific Inverse Problems

Nouamane Laanait, Joshua Romero, Junqi Yin, M. Todd Young, Sean Treichler, Vitalii Starchenko, Albina Borisevich, Alex Sergeev, Michael Matheson

View

Download (PDF)

Source codes

Tags: Computer science, Deep learning, Neural networks, nVidia, Package, Tesla V100

September 29, 2019 by hgpu

Futhark Vulkan Backend

Steffen Holst Larsen

View

Download (PDF)

Source codes

Tags: Computer science, nVidia, nVidia GeForce GTX 780 Ti, OpenCL, Package, Vulkan

September 29, 2019 by hgpu

Heterogeneous Resource-Elastic Management for FPGAs: Concepts, Theory and Implementation

Anuj Vaishnav, Dirk Koch, Khoa Dang Pham

View

Download (PDF)

Tags: Computer science, FPGA, Heterogeneous systems, HLS, OpenCL

September 29, 2019 by hgpu

Elastic deep learning in multi-tenant GPU cluster

Yidi Wu, Kaihao Ma, Xiao Yan, Zhi Liu, James Cheng

View

Download (PDF)

Tags: Computer science, Deep learning, Distributed computing, GPU cluster, nVidia, nVidia GeForce GTX 1080 Ti, Tesla V100

September 29, 2019 by hgpu

ALBERT: A Lite BERT for Self-supervised Learning of Language Representations

Zhenzhong Lan, Mingda Chen, Sebastian Goodman, Kevin Gimpel, Piyush Sharma, Radu Soricut

View

Download (PDF)

Tags: Artificial intelligence, Computer science, Deep learning, NLP, TPU

September 29, 2019 by hgpu

Megatron-LM: Training Multi-Billion Parameter Language Models Using Model Parallelism

Mohammad Shoeybi, Mostofa Patwary, Raul Puri, Patrick LeGresley, Jared Casper, Bryan Catanzaro

View

Download (PDF)

Source codes

Tags: Computer science, Deep learning, NLP, nVidia, nVidia DGX-2, Package, Tesla V100

September 22, 2019 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

Applications

hlslib: Software Engineering for Hardware Design

Performance Impact of Memory Channels on Sparse and Irregular Algorithms

Taichi: A Language for High-Performance Computation on Spatially Sparse Data Structures

Verification of GPU Program Optimizations in Lean

Syntix: A Profiling Based Resource Estimator for CUDA Kernels

MIOpen: An Open Source Library For Deep Learning Primitives

Exascale Deep Learning for Scientific Inverse Problems

Futhark Vulkan Backend

Heterogeneous Resource-Elastic Management for FPGAs: Concepts, Theory and Implementation

Elastic deep learning in multi-tenant GPU cluster

ALBERT: A Lite BERT for Self-supervised Learning of Language Representations

Megatron-LM: Training Multi-Billion Parameter Language Models Using Model Parallelism

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)