high performance computing on graphics processing units: hgpu.org

Charles Weill, Javier Gonzalvo, Vitaly Kuznetsov, Scott Yang, Scott Yak, Hanna Mazzawi, Eugen Hotaj, Ghassen Jerfel, Vladimir Macko, Ben Adlam, Mehryar Mohri, Corinna Cortes

View

Download (PDF)

Source codes

Tags: Algorithms, Computer science, Machine learning, Neural networks, nVidia, Package, TensorFlow, Tesla V100

May 5, 2019 by hgpu

Transparent Acceleration for Heterogeneous Platforms With Compilation to OpenCL

Heinrich Riebler, Gavin Vaz, Tobias Kenter, Christian Plessl

View

Download (PDF)

Tags: Compilers, Computer science, Heterogeneous systems, Intel Xeon Phi, nVidia, OpenCL, Tesla K20

May 1, 2019 by hgpu

The Risks of WebGL: Analysis, Evaluation and Detection

Alex Belkin, Nethanel Gelernter, Israel Cidon

View

Download (PDF)

Tags: Computer science, nVidia, nVidia GeForce GTX 1080, Security, WebGL

May 1, 2019 by hgpu

A mechanism for balancing accuracy and scope in cross-machine black-box GPU performance modeling

James D. Stevens

View

Download (PDF)

Tags: AMD Radeon R9 Fury, ATI, Code generation, Computer science, Finite difference, Heterogeneous systems, Matrix multiplication, nVidia, nVidia GeForce GTX Titan X, OpenCL, Performance, Tesla C2070, Tesla K40

April 28, 2019 by hgpu

Chunkflow: Distributed Hybrid Cloud Processing of Large 3D Images by Convolutional Nets

Jingpeng Wu, William M. Silversmith, H. Sebastian Seung

View

Download (PDF)

Source codes

Tags: Cloud, CNN, Computer science, CUDA, Deep learning, Electron microscopy, Image processing, Microscopy, nVidia, nVidia GeForce GTX 970, Package, Tesla K80

April 28, 2019 by hgpu

Wasserstein-Fisher-Rao Document Distance

Zihao Wang, Datong Zhou, Yong Zhang, Hao Wu, Chenglong Bao

View

Download (PDF)

Tags: Computer science, Machine learning, Nearest neighbour, nVidia, nVidia GeForce GTX 1080 Ti

April 28, 2019 by hgpu

GPU-based Efficient Join Algorithms on Hadoop

Hongzhi Wang, Ning Li, Zheng Wang, Jianing Li

View

Download (PDF)

Tags: Algorithms, Computer science, CUDA, Databases, Hadoop, nVidia

April 28, 2019 by hgpu

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

Agentic Code Optimization via Compiler-LLM Cooperation

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

* * *

high performance computing on graphics processing units: hgpu.org

Predictable GPGPU Computing in DNN-Driven Autonomous Systems

Performance Engineering for a Tall & Skinny Matrix Multiplication Kernel on GPUs

Charactering and Detecting CUDA Program Bugs

Principles, Techniques, and Tools for Explicit and Automatic Parallelization

Compressed Learning of Deep Neural Networks for OpenCL-Capable Embedded Systems

AdaNet: A Scalable and Flexible Framework for Automatically Learning Ensembles

Transparent Acceleration for Heterogeneous Platforms With Compilation to OpenCL

The Risks of WebGL: Analysis, Evaluation and Detection

A mechanism for balancing accuracy and scope in cross-machine black-box GPU performance modeling

Wasserstein-Fisher-Rao Document Distance

GPU-based Efficient Join Algorithms on Hadoop

Recent source codes

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

Device Virtual Machine (DVM)

Agentic Code Optimization via Compiler-LLM Cooperation

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

KernelSkill: A Multi-Agent Framework for GPU Kernel Optimization

Most viewed papers (last 30 days)