high performance computing on graphics processing units: hgpu.org

hgpu.org » nVidia

LUDA: Boost LSM Key Value Store Compactions with GPUs

Peng Xu, Jiguang Wan, Ping Huang, Xiaogang Yang, Chenlei Tang, Fei Wu, Changsheng Xie

View

Download (PDF)

Tags: Computer science, CUDA, Databases, nVidia, nVidia GeForce GTX 1080 Ti

April 12, 2020 by hgpu

Deep Learning for Compilers

Chris Cummins

View

Download (PDF)

Source codes

Tags: AMD Radeon HD 7970, ATI, Benchmarking, Compilers, Computer science, Deep learning, Heterogeneous systems, nVidia, nVidia GeForce GTX 970, OpenCL, Package, Thesis

April 5, 2020 by hgpu

Parallelization of the Honeybee Search Algorithm for Object Tracking

Oscar E. Perez-Cham, Cesar Puente, Carlos Soubervielle-Montalvo, Gustavo Olague, Carlos A. Aguirre-Salado, Alberto S. Nuñez-Varela

View

Download (PDF)

Tags: AMD Radeon R9 270, Artificial intelligence, ATI, Computer science, Computer vision, nVidia, nVidia GeForce GTX 1050, OpenCL, Particle swarm optimization

April 5, 2020 by hgpu

PyMatting: A Python Library for Alpha Matting

Thomas Germer, Tobias Uelwer, Stefan Conrad, Stefan Harmeling

View

Download (PDF)

Source codes

Tags: Computer vision, CUDA, Image processing, nVidia, OpenCL, Package, PyOpenCL, Python

April 5, 2020 by hgpu

Faster than FAST: GPU-Accelerated Frontend for High-Speed VIO

Balazs Nagy, Philipp Foehn, Davide Scaramuzza

View

Download (PDF)

Source codes

Tags: Algorithms, Computer vision, CUDA, Image processing, nVidia, nVidia GeForce GTX 960 M, nVidia Jetson TX2, OpenCV, Package

April 5, 2020 by hgpu

Understanding GPU-Based Lossy Compression for Extreme-Scale Cosmological Simulations

Sian Jin, Pascal Grosset, Christopher M. Biwer, Jesus Pulido, Jiannan Tian, Dingwen Tao, James Ahrens

View

Download (PDF)

Source codes

Tags: Astrophysics, Compression, Computational Physics, Cosmology, CUDA, Instrumentation and Methods for Astrophysics, nVidia, nVidia GeForce GTX 1080 Ti, nVidia GeForce GTX Titan V, nVidia GeForce RTX 2080 Ti, nVidia Quadro P 6000, Package, Physics, Tesla K80, Tesla P100, Tesla V100

April 5, 2020 by hgpu

Large-Scale Data Computing Performance Comparisons on SYCL Heterogeneous Parallel Processing Layer Implementations

Woosuk Shin, Kwan-Hee Yoo, Nakhoon Baek

View

Download (PDF)

Tags: Computer science, CUDA, DSL, Heterogeneous systems, nVidia, nVidia GeForce GTX 960, OpenCL, SYCL

March 29, 2020 by hgpu

Characterizing Optimizations to Memory Access Patterns using Architecture-Independent Program Features

Aditya Chilukuri, Josh Milthorpe, Beau Johnston

View

Download (PDF)

Tags: Computer science, Heterogeneous systems, Memory, nVidia, OpenCL, Performance, Tesla P100

March 29, 2020 by hgpu

SOL: Effortless Device Support for AI Frameworks without Source Code Changes

Nicolas Weber, Felipe Huici

View

Download (PDF)

Tags: Artificial intelligence, Computer science, CUDA, Deep learning, Heterogeneous systems, nVidia, nVidia GeForce GTX Titan V100, nVidia Quadro P 4000, PyTorch

March 29, 2020 by hgpu

ProGraML: Graph-based Deep Learning for Program Optimization and Analysis

Chris Cummins, Zacharias V. Fisches, Tal Ben-Nun, Torsten Hoefler, Hugh Leather

View

Download (PDF)

Source codes

Tags: Computer science, Deep learning, Machine learning, nVidia, nVidia GeForce GTX 1080, nVidia GeForce GTX 970, OpenCL, Package, Performance, Programming Languages

March 29, 2020 by hgpu

ELECTRA: Pre-training Text Encoders as Discriminators Rather Than Generators

Kevin Clark, Minh-Thang Luong, Quoc V. Le, Christopher D. Manning

View

Download (PDF)

Source codes

Tags: Computer science, NLP, nVidia, Package, Tesla V100, TPU

March 29, 2020 by hgpu

Fireiron: A Scheduling Language for High-Performance Linear Algebra on GPUs

Bastian Hagedorn, Archibald Samuel Elliott, Henrik Barthels, Rastislav Bodik, Vinod Grover

View

Download (PDF)

Tags: Computer science, CUBLAS, CUDA, Linear Algebra, Matrix multiplication, nVidia, nVidia GeForce GTX 750 Ti, nVidia GeForce RTX 2080 Ti, nVidia Quadro GV100, Programming Languages

March 22, 2020 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

LUDA: Boost LSM Key Value Store Compactions with GPUs

Deep Learning for Compilers

Parallelization of the Honeybee Search Algorithm for Object Tracking

PyMatting: A Python Library for Alpha Matting

Faster than FAST: GPU-Accelerated Frontend for High-Speed VIO

Understanding GPU-Based Lossy Compression for Extreme-Scale Cosmological Simulations

Large-Scale Data Computing Performance Comparisons on SYCL Heterogeneous Parallel Processing Layer Implementations

Characterizing Optimizations to Memory Access Patterns using Architecture-Independent Program Features

SOL: Effortless Device Support for AI Frameworks without Source Code Changes

ProGraML: Graph-based Deep Learning for Program Optimization and Analysis

ELECTRA: Pre-training Text Encoders as Discriminators Rather Than Generators

Fireiron: A Scheduling Language for High-Performance Linear Algebra on GPUs

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)