high performance computing on graphics processing units: hgpu.org

Programming

hgpu.org » Programming

Memory layout in GPU implementation of lattice Boltzmann method for sparse 3D geometries

Tadeusz Tomczak, Roman G. Szafran

View

Download (PDF)

Tags: cfd, CUDA, Fluid dynamics, Lattice Boltzmann model, nVidia, nVidia GeForce GTX Titan

November 10, 2016 by hgpu

Balancing locality and concurrency: solving sparse triangular systems on GPUs

Andrea Picciau, Gordon E. Inggs, John Wickerson, Eric C. Kerrigan, George A. Constantinides

View

Download (PDF)

Tags: Algorithms, AMD FirePro W5000, ATI, Computer science, Linear Algebra, nVidia, nVidia Quadro K4000, OpenCL, Sparse matrix

November 8, 2016 by hgpu

Tamp: A Library for Compact Deep Neural Networks with Structured Matrices

Bingchen Gong, Brendan Jou, Felix Yu, Shih-Fu Chang

View

Download (PDF)

Source codes

Tags: Caffe, Computer science, CUDA, Deep learning, Neural networks, nVidia, Package, TensorFlow

November 8, 2016 by hgpu

Performance Portability of the Aeras Atmosphere Model to Next Generation Architectures using Kokkos

Jerry Watkins, Irina Tezaur

View

Download (PDF)

Tags: CUDA, Earth and Space Sciences, MPI, nVidia, OpenMP, performance portability, Physics, Tesla K20, Tesla K80

November 8, 2016 by hgpu

Vispark: GPU-Accelerated Distributed Visual Computing Using Spark

Woohyuk Choi, Sumin Hong, Won-Ki Jeong

View

Download (PDF)

Tags: Clustering, Computer science, CUDA, Distributed computing, DSL, GPU cluster, Image processing, MapReduce, nVidia, nVidia GeForce GTX Titan, Python, Rendering, Spark

November 8, 2016 by hgpu

UNICORN: A Bulk Synchronous Programming Model, Framework and Runtime for Hybrid CPU-GPU Clusters

Tarun Beri

View

Download (PDF)

Source codes

Tags: Computer science, CUDA, FFT, GPU cluster, Linear Algebra, Matrix multiplication, nVidia, Package, Prefetch, Tesla M2070, Thesis

November 5, 2016 by hgpu

Molecular Activity Prediction using Deep Learning Software Library

Yoshiki Kato, Shinji Hamada, Hitoshi Goto

View

Download (PDF)

Source codes

Tags: Bioinformatics, Chemistry, Computer science, CUDA, Deep learning, nVidia, Package, Python, Tesla K20

November 5, 2016 by hgpu

HPVM: A Portable Virtual Instruction Set for Heterogeneous Parallel Systems

Prakalp Srivastava, Maria Kotsifakou, Vikram Adve

View

Download (PDF)

Tags: Code generation, Computer science, Data parallelism, Heterogeneous systems, nVidia, nVidia GeForce GTX 680, OpenCL, performance portability

November 5, 2016 by hgpu

A Memory Bandwidth-Efficient Hybrid Radix Sort on GPUs

Elias Stehle, Hans-Arno Jacobsen

View

Download (PDF)

Tags: Algorithms, Computer science, CUDA, Heterogeneous systems, nVidia, nVidia GeForce GTX Titan X, Sorting

November 5, 2016 by hgpu

grim: A Flexible, Conservative Scheme for Relativistic Fluid Theories

Mani Chandra, Francois Foucart, Charles F. Gammie

View

Download (PDF)

Source codes

Tags: Astrophysics, Magnetohydrodynamics, nVidia, OpenCL, Package, Physics, Tesla K20

November 5, 2016 by hgpu

Extensions and Limitations of the Neural GPU

Eric Price, Wojciech Zaremba, Ilya Sutskever

View

Download (PDF)

Source codes

Tags: Algorithms, Computer science, CUDA, Deep learning, nVidia, Package, Python

November 3, 2016 by hgpu

Diplomat: Mapping of multi-kernel applications using a static dataflow abstraction

Bruno Bodin, Luigi Nardi, Paul H. J. Kelly, Michael F. P. O'Boyle

View

Download (PDF)

Tags: ARM, Computer science, Computer vision, Heterogeneous systems, OpenCL

November 3, 2016 by hgpu

KernelSkill: A Multi-Agent Framework for GPU Kernel Optimization

MSKernelBench & CUDAMaster

Making LLMs Optimize Multi-Scenario CUDA Kernels Like Experts

EvoScientist: Harness Vibe Research with Self-evolving AI Scientists

EvoScientist: Towards Multi-Agent Evolving AI Scientists for End-to-End Scientific Discovery

RepoLaunch: Automating Build and Test Pipeline of Code Repositories on ANY Language and ANY Platform

RepoLaunch: Automating Build & Test Pipeline of Code Repositories on ANY Language and ANY Platform

RepoLaunch: Automating Build and Test Pipeline of Code Repositories on ANY Language and ANY Platform

RepoLaunch: Automating Build & Test Pipeline of Code Repositories on ANY Language and ANY Platform

CONCUR: a benchmark designed to evaluate multithreaded Java code generated by LLMs

CONCUR: Benchmarking LLMs for Concurrent Code Generation

HIPRT: Ray Tracing using HIP

Ray Tracing using HIP

MXFP4 Training Support Codebase

Practical FP4 Training for Large-Scale MoE Models on Hopper GPUs

CUDA Agent: Large-Scale Agentic RL for High-Performance CUDA Kernel Generation

CUDABench: Benchmarking LLMs for Text-to-CUDA Generation

See all packages

* * *

high performance computing on graphics processing units: hgpu.org

Programming

Memory layout in GPU implementation of lattice Boltzmann method for sparse 3D geometries

Balancing locality and concurrency: solving sparse triangular systems on GPUs

Tamp: A Library for Compact Deep Neural Networks with Structured Matrices

Performance Portability of the Aeras Atmosphere Model to Next Generation Architectures using Kokkos

Vispark: GPU-Accelerated Distributed Visual Computing Using Spark

UNICORN: A Bulk Synchronous Programming Model, Framework and Runtime for Hybrid CPU-GPU Clusters

Molecular Activity Prediction using Deep Learning Software Library

HPVM: A Portable Virtual Instruction Set for Heterogeneous Parallel Systems

A Memory Bandwidth-Efficient Hybrid Radix Sort on GPUs

grim: A Flexible, Conservative Scheme for Relativistic Fluid Theories

Extensions and Limitations of the Neural GPU

Diplomat: Mapping of multi-kernel applications using a static dataflow abstraction

Recent source codes

KernelSkill: A Multi-Agent Framework for GPU Kernel Optimization

MSKernelBench & CUDAMaster

EvoScientist: Harness Vibe Research with Self-evolving AI Scientists

RepoLaunch: Automating Build and Test Pipeline of Code Repositories on ANY Language and ANY Platform

RepoLaunch: Automating Build and Test Pipeline of Code Repositories on ANY Language and ANY Platform

CONCUR: a benchmark designed to evaluate multithreaded Java code generated by LLMs

HIPRT: Ray Tracing using HIP

MXFP4 Training Support Codebase

CUDA Agent: Large-Scale Agentic RL for High-Performance CUDA Kernel Generation

CUDABench: Benchmarking LLMs for Text-to-CUDA Generation

Most viewed papers (last 30 days)