high performance computing on graphics processing units: hgpu.org

Akshay Sridhar, Yassine Tissaoui, Simone Marras, Zhaoyi Shen, Charles Kawczynski, Simon Byrne, Kiran Pamnany, Maciej Waruszewski, Thomas H. Gibson, Jeremy E. Kozdon, Valentin Churavy, Lucas C. Wilcox, Francis X. Giraldo, Tapio Schneider

View

Download (PDF)

Source codes

Tags: CUDA, Earth and Space Sciences, nVidia, Package, Tesla V100, Weather prediction

October 10, 2021 by hgpu

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

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

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

* * *

high performance computing on graphics processing units: hgpu.org

Programming

Bolt: Bridging the Gap between Auto-tuners and Hardware-native Performance

The CoRa Tensor Compiler: Compilation for Ragged Tensors with Minimal Padding

Monitoring Collective Communication Among GPUs

Principal Kernel Analysis: A Tractable Methodology to Simulate Scaled GPU Workloads

OMB-Py: Python Micro-Benchmarks for Evaluating Performance of MPI Libraries on HPC Systems

Least Squares on GPUs in Multiple Double Precision

Homomorphic-Encrypted Volume Rendering

Accelerating LBM on a Tightly-Coupled Field Programmable Gate Array

Accelerating AutoDock VINA with GPUs

Beyond Desktop Computation: Challenges in Scaling a GPU Infrastructure

Artificial Intelligence in Electric Machine Drives: Advances and Trends

Large-eddy simulations with ClimateMachine: a new open-source code for atmospheric simulations on GPUs and CPUs

Recent source codes

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

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

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

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

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)