high performance computing on graphics processing units: hgpu.org

hgpu.org » Code generation

Autotuning, Code Generation and Optimizing Compiler Technology for GPUs

Malik Muhammad Zaki Murtaza Khan

View

Download (PDF)

Tags: Code generation, Computer science, CUBLAS, CUDA, Linear Algebra, nVidia, Tesla C2050, Thesis

January 26, 2013 by hgpu

Interactive Refactoring for GPU Parallelization of Affine Loops

Kostadin Damevski, Madhan Muralimanohar

View

Download (PDF)

Tags: Code generation, Computer science, CUDA, Heterogeneous systems, nVidia, nVidia Quadro FX 2000

January 7, 2013 by hgpu

Automatic Code Generation for Stencil Computations on GPU Architectures

Justin Andrew Holewinski

View

Download (PDF)

Tags: Code generation, Computer science, CUDA, nVidia, nVidia GeForce GTX 280, OpenCL, Tesla C2050, Tesla K10, Thesis

January 4, 2013 by hgpu

Towards Domain-specific Computing for Stencil Codes in HPC

Richard Membarth, Frank Hannig, Jurgen Teich, Harald Kostler

View

Download (PDF)

Source codes

Tags: Algorithms, Code generation, Compression, Computer science, CUDA, Heterogeneous systems, nVidia, nVidia Quadro FX 5800, OpenCL, Package, Tesla C2050

December 12, 2012 by hgpu

Lightweight Modular Staging and Embedded Compilers: Abstraction Without Regret for High-Level High-Performance Programming

Tiark Rompf

View

Download (PDF)

Tags: Code generation, Compilers, Computer science, CUDA, Heterogeneous systems, nVidia, Performance, Software Engineering, Tesla C2050, Thesis

December 8, 2012 by hgpu

Developing Extensible Lattice-Boltzmann Simulators for General-Purpose Graphics-Processing Units

Stuart D. C. Walsh, Martin O. Saar

View

Download (PDF)

Tags: Code generation, CUDA, Fluid dynamics, Lattice Boltzmann model, nVidia, Tesla C1060

December 1, 2012 by hgpu

A compiler toolkit for array-based languages targeting CPU/GPU hybrid systems

Rahul Garg, Laurie Hendren

View

Download (PDF)

Source codes

Tags: ATI, ATI Radeon HD 5850, ATI Radeon HD 7970, Code generation, Compilers, Computer science, OpenCL, Package, Python, Tesla C2050

November 26, 2012 by hgpu

Auto-tuning on the macro scale: high level algorithmic auto-tuning for scientific applications

Cy P Chan

View

Download (PDF)

Tags: Algorithms, Code generation, Computer science, nVidia, nVidia GeForce GTX 280, Optimization, Thesis

November 23, 2012 by hgpu

Sigma*: Symbolic Learning of Input-Output Specifications

Matko Botincan, Domagoj Babic

View

Download (PDF)

Tags: Code generation, Computer science, CUDA, nVidia, nVidia GeForce GTX 460, Programming Languages

November 10, 2012 by hgpu

GPU Virtualization

Kristoffer Robin Stokke

View

Download (PDF)

Source codes

Tags: Code generation, Computer science, CUDA, nVidia, nVidia GeForce GTX 280, nVidia Quadro NVS 295, Package, Python, Thesis, Virtualization

November 7, 2012 by hgpu

An Intermediate Library for Multi-GPUs Computing Skeletons

Le Duc Tung, Nguyen Huu Duc, Pham Tuan Anh, Ngo Huy Hoang, Nguyen Minh Thap

View

Download (PDF)

Tags: Code generation, Computer science, CUDA, GPU cluster, nVidia, nVidia GeForce GTX 295

November 1, 2012 by hgpu

Mapping a Dataflow Programming Model onto Heterogeneous Architectures

Alina Sbirlea

View

Download (PDF)

Tags: Code generation, Computer science, CUDA, Heterogeneous systems, nVidia, nVidia Quadro FX 580, Tesla C1060, Thesis

September 12, 2012 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

Autotuning, Code Generation and Optimizing Compiler Technology for GPUs

Interactive Refactoring for GPU Parallelization of Affine Loops

Automatic Code Generation for Stencil Computations on GPU Architectures

Towards Domain-specific Computing for Stencil Codes in HPC

Lightweight Modular Staging and Embedded Compilers: Abstraction Without Regret for High-Level High-Performance Programming

Developing Extensible Lattice-Boltzmann Simulators for General-Purpose Graphics-Processing Units

A compiler toolkit for array-based languages targeting CPU/GPU hybrid systems

Auto-tuning on the macro scale: high level algorithmic auto-tuning for scientific applications

Sigma*: Symbolic Learning of Input-Output Specifications

GPU Virtualization

An Intermediate Library for Multi-GPUs Computing Skeletons

Mapping a Dataflow Programming Model onto Heterogeneous Architectures

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)