high performance computing on graphics processing units: hgpu.org

Programming

hgpu.org » Programming

Out of kernel tuning and optimizations for portable large-scale docking experiments on GPUs

Gianmarco Accordi, Davide Gadioli, Emanele Vitali, Luigi Crisci, Biagio Cosenza, Andrea Beccari, Gianluca Palermo

View

Download (PDF)

Tags: Computer science, CUDA, HPC, molecular docking, nVidia, nVidia A100, oneAPI, Performance, SYCL

February 12, 2024 by hgpu

Training DNN Models over Heterogeneous Clusters with Optimal Performance

Chengyi Nie, Jessica Maghakian, Zhenhua Liu

View

Download (PDF)

Tags: Computer science, CUDA, Deep learning, Heterogeneous systems, load balancing, Neural networks, nVidia, nVidia RTX 6000 Ada

February 12, 2024 by hgpu

Gallatin: A General-Purpose GPU Memory Manager

Hunter McCoy, Prashant Pandey

View

Download (PDF)

Source codes

Tags: Computer science, CUDA, HPC, Memory, nVidia, nVidia A40, Package

February 4, 2024 by hgpu

Deductive verification for SYCL

Ellen Wittingen

View

Download (PDF)

Source codes

Tags: Computer science, CUDA, Heterogeneous systems, HIP, nVidia, OpenCL, OpenMP, Package, SYCL, Thesis

February 4, 2024 by hgpu

LeftoverLocals: Listening to LLM Responses Through Leaked GPU Local Memory

Tyler Sorensen, Heidy Khlaaf

View

Download (PDF)

Source codes

Tags: Computer science, Data recovery, nVidia, nVidia GeForce RTX 4070, OpenCL, Package, Security, Vulkan

February 4, 2024 by hgpu

Towards a GPU-Parallelization of the neXtSIM-DG Dynamical Core

Robert Jendersie, Christian Lessig, Thomas Richter

View

Download (PDF)

Source codes

Tags: Computer science, CUDA, Earth and Space Sciences, Finite element method, Numerical Analysis, nVidia, nVidia A100, nVidia GeForce RTX 3090, OpenMP, Package, PyTorch, SYCL

February 4, 2024 by hgpu

High-order thread-safe lattice Boltzmann model for HPC turbulent flow simulations

Andrea Montessori, Michele La Rocca, Giorgio Amati, Marco Lauricella, Adriano Tiribocchi, Sauro Succi

View

Download (PDF)

Source codes

Tags: cfd, CUDA, Fluid dynamics, HPC, Lattice Boltzmann model, nVidia, nVidia A100, nVidia GeForce RTX 3090, Package

February 4, 2024 by hgpu

Assessing the Impact of Compiler Optimizations on GPUs Reliability

Fernando Fernandes dos Santos, Luigi Carro, Flavio Vella, Paolo Rech

View

Download (PDF)

Tags: Compilers, Computer science, CUDA, nVidia, nVidia GeForce GTX Titan V, PTX, Tesla K40, Tesla V100

January 28, 2024 by hgpu

Lessons Learned Migrating CUDA to SYCL: A HEP Case Study with ROOT RDataFrame

Jolly Chen, Monica Dessole, Ana Lucia Varbanescu

View

Download (PDF)

Source codes

Tags: CUDA, HEP, nVidia, nVidia Quadro RTX 4000, OpenCL, Package, Performance, Physics, SYCL

January 28, 2024 by hgpu

A Heterogeneous Inference Framework for a Deep Neural Network

Rafael Gadea-Gironés, José Luís Rocabado-Rocha, Jorge Fe, Jose M. Monzo

View

Download (PDF)

Tags: Artificial intelligence, Computer science, Deep learning, FPGA, Heterogeneous systems, HLS, Machine learning, Neural networks, OpenCL, PyTorch

January 28, 2024 by hgpu

BANG: Billion-Scale Approximate Nearest Neighbor Search using a Single GPU

Karthik V., Saim Khan, Somesh Singh, Harsha Vardhan Simhadri, Jyothi Vedurada

View

Download (PDF)

Tags: Computer science, CUDA, Nearest neighbour, nVidia, nVidia A100

January 28, 2024 by hgpu

swCUDA: Auto parallel code translation framework from CUDA to ATHREAD for new generation sunway supercomputer

Maoxue Yu, Guanghao Ma, Zhuoya Wang, Shuai Tang, Yuhu Chen, Yucheng Wang, Yuanyuan Liu, Dongning Jia, Zhiqiang Wei

View

Download (PDF)

Tags: Code generation, Computer science, CUDA, nVidia, nVidia GeForce GTX 1050 Ti

January 21, 2024 by hgpu

CuTile Benchmark Suite: Performance and Productivity Tradeoffs for GPU Kernel Programming on Blackwell Architecture

Evaluating CUDA Tile for AI Workloads on Hopper and Blackwell GPUs

Agentic Code Optimization via Compiler-LLM Cooperation

MegaTrain: Full Precision Training of 100B+ Parameter Large Language Models on a Single GPU

Device Virtual Machine (DVM)

DVM: Real-Time Kernel Generation for Dynamic AI Models

AutoKernel: Autoresearch for GPU kernels. Give it any PyTorch model, go to sleep, wake up to optimized Triton kernels

AutoKernel: Autonomous GPU Kernel Optimization via Iterative Agent-Driven Search

Triton-Sanitizer: A Fast and Device-Agnostic Memory Sanitizer for Triton with Rich Diagnostic Context

LLM.Q: Quantized LLM training in pure CUDA/C++

LLMQ: Efficient Lower-Precision LLM Training for Consumer GPUs

SOL-ExecBench: Speed-of-Light Benchmarking for Real-World GPU Kernels Against Hardware Limits

True 4-Bit Quantized CNN Training on CPU

True 4-Bit Quantized Convolutional Neural Network Training on CPU: Achieving Full-Precision Parity

cuFuzz: A GPU-oriented coverage-guided fuzzer for userland CUDA application

Hunting CUDA Bugs at Scale with cuFuzz

See all packages

* * *

high performance computing on graphics processing units: hgpu.org

Programming

Out of kernel tuning and optimizations for portable large-scale docking experiments on GPUs

Training DNN Models over Heterogeneous Clusters with Optimal Performance

Gallatin: A General-Purpose GPU Memory Manager

Deductive verification for SYCL

LeftoverLocals: Listening to LLM Responses Through Leaked GPU Local Memory

Towards a GPU-Parallelization of the neXtSIM-DG Dynamical Core

High-order thread-safe lattice Boltzmann model for HPC turbulent flow simulations

Assessing the Impact of Compiler Optimizations on GPUs Reliability

Lessons Learned Migrating CUDA to SYCL: A HEP Case Study with ROOT RDataFrame

A Heterogeneous Inference Framework for a Deep Neural Network

BANG: Billion-Scale Approximate Nearest Neighbor Search using a Single GPU

swCUDA: Auto parallel code translation framework from CUDA to ATHREAD for new generation sunway supercomputer

Recent source codes

CuTile Benchmark Suite: Performance and Productivity Tradeoffs for GPU Kernel Programming on Blackwell Architecture

Agentic Code Optimization via Compiler-LLM Cooperation

MegaTrain: Full Precision Training of 100B+ Parameter Large Language Models on a Single GPU

Device Virtual Machine (DVM)

AutoKernel: Autoresearch for GPU kernels. Give it any PyTorch model, go to sleep, wake up to optimized Triton kernels

Triton-Sanitizer: A Fast and Device-Agnostic Memory Sanitizer for Triton with Rich Diagnostic Context

LLM.Q: Quantized LLM training in pure CUDA/C++

SOL-ExecBench: Speed-of-Light Benchmarking for Real-World GPU Kernels Against Hardware Limits

True 4-Bit Quantized CNN Training on CPU

cuFuzz: A GPU-oriented coverage-guided fuzzer for userland CUDA application

Most viewed papers (last 30 days)