high performance computing on graphics processing units: hgpu.org

hgpu.org » nVidia A100

Breaking the Memory Wall: A Study of I/O Patterns and GPU Memory Utilization for Hybrid CPU-GPU Offloaded Optimizers

Avinash Maurya, Jie Ye, M. Mustafa Rafique, Franck Cappello, Bogdan Nicolae

View

Download (PDF)

Tags: Computer science, CUDA, LLM, Memory, nVidia, nVidia A100, nVidia H100, Performance

June 23, 2024 by hgpu

Optimal Kernel Orchestration for Tensor Programs with Korch

Muyan Hu, Ashwin Venkatram, Shreyashri Biswas, Balamurugan Marimuthu, Bohan Hou, Gabriele Oliaro, Haojie Wang, Liyan Zheng, Xupeng Miao, Jidong Zhai

View

Download (PDF)

Source codes

Tags: Compilers, Computer science, CUDA, Neural networks, nVidia, nVidia A100, nVidia V100, Package, PyTorch

June 23, 2024 by hgpu

Stencil Computations on AMD and Nvidia Graphics Processors: Performance and Tuning Strategies

Johannes Pekkilä, Oskar Lappi, Fredrik Robertsén, Maarit J. Korpi-Lagg

View

Download (PDF)

Source codes

Tags: AMD Radeon Instinct MI100, AMD Radeon Instinct MI250X, ATI, Computer science, CUDA, Energy-efficient computing, HIP, nVidia, nVidia A100, nVidia V100, Package, Performance, PyTorch, Stencil computation

June 16, 2024 by hgpu

How much can we gain from Tensor Kernel Fusion on GPUs?

Wei Sun, Ang Li, Sander Stuijk, Henk Corporaal

View

Download (PDF)

Tags: Computer science, CUDA, Deep learning, Matrix multiplication, Neural networks, nVidia, nVidia A100, nVidia H100

June 16, 2024 by hgpu

Gaining Cross-Platform Parallelism for HAL’s Molecular Dynamics Package using SYCL

Viktor Skoblin, Felix Höfling, Steffen Christgau

View

Download (PDF)

Source codes

Tags: AMD Radeon Instinct MI210, ATI, Computer science, CUDA, Molecular dynamics, nVidia, nVidia A100, nVidia A40, Package, Physics, SYCL

June 9, 2024 by hgpu

Helix: Distributed Serving of Large Language Models via Max-Flow on Heterogeneous GPUs

Yixuan Mei, Yonghao Zhuang, Xupeng Miao, Juncheng Yang, Zhihao Jia, Rashmi Vinayak

View

Download (PDF)

Tags: Computer science, GPU cluster, Heterogeneous systems, nVidia, nVidia A100, nVidia V100, Tesla T4

June 9, 2024 by hgpu

Evaluation of computational and energy performance in matrix multiplication algorithms on CPU and GPU using MKL, cuBLAS and SYCL

L.A. Torres, Carlos J. Barrios H, Yves Denneulin

View

Download (PDF)

Source codes

Tags: Computer science, CUBLAS, CUDA, Linear Algebra, Matrix multiplication, Neural networks, nVidia, nVidia A100, Package, Performance, SYCL

June 2, 2024 by hgpu

An implementation of tensor product patch smoothers on GPU

Cu Cui, Paul Grosse-Bley, Guido Kanschat, Robert Strzodka

View

Download (PDF)

Tags: CUDA, FEM, Finite element method, Mathematics, Numerical Analysis, nVidia, nVidia A100

June 2, 2024 by hgpu

Enabling full-speed random access to the entire memory on the A100 GPU

Alden Walker

View

Download (PDF)

Tags: Computer science, nVidia, nVidia A100, Performance, PTX

May 26, 2024 by hgpu

ArchesWeather: An efficient AI weather forecasting model at 1.5° resolution

Guillaume Couairon, Christian Lessig, Anastase Charantonis, Claire Monteleoni

View

Download (PDF)

Tags: Deep learning, Earth and Space Sciences, nVidia, nVidia A100, Weather prediction

May 26, 2024 by hgpu

GPU Implementations for Midsize Integer Addition and Multiplication

Cosmin E. Oancea, Stephen M. Watt

View

Download (PDF)

Tags: Algorithms, Computer science, CUDA, nVidia, nVidia A100, Performance, Programming Languages

May 26, 2024 by hgpu

STuning-DL: Model-Driven Autotuning of Sparse GPU Kernels for Deep Learning

Roberto L. Castro, Diego Andrade, Basilio B. Fraguela

View

Download (PDF)

Tags: Auto-Tuning, Computer science, CUDA, Deep learning, Machine learning, nVidia, nVidia A100, Tesla T4

May 26, 2024 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

Breaking the Memory Wall: A Study of I/O Patterns and GPU Memory Utilization for Hybrid CPU-GPU Offloaded Optimizers

Optimal Kernel Orchestration for Tensor Programs with Korch

Stencil Computations on AMD and Nvidia Graphics Processors: Performance and Tuning Strategies

How much can we gain from Tensor Kernel Fusion on GPUs?

Gaining Cross-Platform Parallelism for HAL’s Molecular Dynamics Package using SYCL

Helix: Distributed Serving of Large Language Models via Max-Flow on Heterogeneous GPUs

Evaluation of computational and energy performance in matrix multiplication algorithms on CPU and GPU using MKL, cuBLAS and SYCL

An implementation of tensor product patch smoothers on GPU

Enabling full-speed random access to the entire memory on the A100 GPU

ArchesWeather: An efficient AI weather forecasting model at 1.5° resolution

GPU Implementations for Midsize Integer Addition and Multiplication

STuning-DL: Model-Driven Autotuning of Sparse GPU Kernels for Deep Learning

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)