high performance computing on graphics processing units: hgpu.org

hgpu.org » LLVM

Abstractions for Programming Graphics Processors in High-Level Programming Languages

Tim Besard

View

Download (PDF)

Source codes

Tags: Computer science, CUDA, Heterogeneous systems, High-level Languages, LLVM, nVidia, nVidia GeForce GTX 1080, Package, Thesis

December 29, 2019 by hgpu

KLARAPTOR: A Tool for Dynamically Finding Optimal Kernel Launch Parameters Targeting CUDA Programs

Alexander Brandt, Davood Mohajerani, Marc Moreno Maza, Jeeva Paudel, Linxiao Wang

View

Download (PDF)

Source codes

Tags: Computer science, CUDA, LLVM, nVidia, Package, Performance

November 10, 2019 by hgpu

WCCV: Improving the Vectorization of IF-statements with Warp-Coherent Conditions

Huihui Sun, Florian Fey, Jie Zhao, Sergei Gorlatch

View

Download (PDF)

Source codes

Tags: Benchmarking, Code generation, Computer science, LLVM, OpenCL, Package, Performance

June 30, 2019 by hgpu

Triton: An Intermediate Language and Compiler for Tiled Neural Network Computations

Philippe Tillet, H. T. Kung, David Cox

View

Download (PDF)

Source codes

Tags: Compilers, Computer science, CUDA, Deep learning, High-level Languages, LLVM, Matrix multiplication, Neural networks, nVidia, nVidia GeForce GTX 1070, Package

June 30, 2019 by hgpu

An optimizing multi-platform source-to-source compiler framework for the NEURON MODeling Language

Pramod Kumbhar, Omar Awile, Liam Keegan, Jorge Blanco Alonso, James King, Michael Hines, Felix Schuermann

View

Download (PDF)

Source codes

Tags: Biophysics, Code generation, Compilers, Computer science, CUDA, HPC, Intel Xeon Phi, LLVM, nVidia, OpenACC, OpenMP, Package

May 15, 2019 by hgpu

Charactering and Detecting CUDA Program Bugs

Mingyuan Wu, Husheng Zhou, Lingming Zhang, Cong Liu, Yuqun Zhang

View

Download (PDF)

Source codes

Tags: Computer science, CUDA, LLVM, nVidia, Package, Software Engineering

May 8, 2019 by hgpu

Performance Evaluation of OpenMP’s Target Construct on GPUs: Exploring Compiler Optimizations

Akihiro Hayashi, Jun Shirako, Etorre Tiotto, Robert Ho, Vivek Sarkar

View

Download (PDF)

Tags: Compilers, Computer science, CUDA, LLVM, nVidia, OpenMP, Optimization, Tesla K80, Tesla P100

February 10, 2019 by hgpu

AutoPhase: Compiler Phase-Ordering for High Level Synthesis with Deep Reinforcement Learning

Ameer Haj-Ali, Qijing Huang, William Moses, John Xiang, Ion Stoica, Krste Asanovic, John Wawrzynek

View

Download (PDF)

Tags: Algorithms, Code generation, Compilers, Computer science, Deep learning, HLS, LLVM, Programming Languages

January 20, 2019 by hgpu

Clacc: Translating OpenACC to OpenMP in Clang

Joel E. Denny, Seyong Lee, Jeffrey S. Vetter

View

Download (PDF)

Tags: Compilers, Computer science, Heterogeneous systems, LLVM, nVidia, nVidia GeForce GTX 1050, OpenACC, OpenMP

December 9, 2018 by hgpu

Performance Evaluation of OpenMP’s Target Construct on GPUs – Exploring Compiler Optimizations

Akihiro Hayashi, Jun Shirako, Etorre Tiotto, Robert Ho, Vivek Sarkar

View

Download (PDF)

Tags: Compilers, Computer science, CUDA, LLVM, nVidia, OpenMP, Performance, Tesla K80, Tesla P100

August 26, 2018 by hgpu

Neural Code Comprehension: A Learnable Representation of Code Semantics

Tal Ben-Nun, Alice Shoshana Jakobovits, Torsten Hoefler

View

Download (PDF)

Tags: ATI, ATI Radeon HD 7970, Code generation, Computer science, CUDA, Deep learning, LLVM, LSTM, Machine learning, NLP, nVidia, nVidia GeForce GTX 970, OpenCL, Programming Languages, RNN

June 20, 2018 by hgpu

OpenCL Vector Swizzling Optimization under Global Value Numbering

Li-An Her, Jenq-Kuen Lee

View

Download (PDF)

Tags: Computer science, Heterogeneous systems, HSA, LLVM, OpenCL

May 20, 2018 by hgpu

MSCCL++: A GPU-driven communication stack for scalable AI applications

MSCCL++: Rethinking GPU Communication Abstractions for Cutting-edge AI Applications

Benchmark compute shader of Unity against InteropUnityCUDA

InteropUnityCUDA: A Tool for Interoperability Between Unity and CUDA

Data-efficient LLM Fine-tuning for Code Generation

DeepSpeed is a deep learning optimization library that makes distributed training and inference easy, efficient, and effective.

DeepCompile: A Compiler-Driven Approach to Optimizing Distributed Deep Learning Training

Large Language Model Powered C-to-CUDA Code Translation: A Novel Auto-Parallelization Framework

GigaAPI: a user-space API that simplifies multi-GPU programming, bridging the gap between the capabilities of parallel GPU systems and the ability of developers to harness their full potential

GigaAPI for GPU Parallelization

Coccinelle: a C code transformation engine using SmPL for matches, refactorings, and bug fixing

Advances in Semantic Patching for HPC-oriented Refactorings with Coccinelle

* * *

high performance computing on graphics processing units: hgpu.org

Abstractions for Programming Graphics Processors in High-Level Programming Languages

KLARAPTOR: A Tool for Dynamically Finding Optimal Kernel Launch Parameters Targeting CUDA Programs

WCCV: Improving the Vectorization of IF-statements with Warp-Coherent Conditions

Triton: An Intermediate Language and Compiler for Tiled Neural Network Computations

An optimizing multi-platform source-to-source compiler framework for the NEURON MODeling Language

Charactering and Detecting CUDA Program Bugs

Performance Evaluation of OpenMP’s Target Construct on GPUs: Exploring Compiler Optimizations

AutoPhase: Compiler Phase-Ordering for High Level Synthesis with Deep Reinforcement Learning

Clacc: Translating OpenACC to OpenMP in Clang

Performance Evaluation of OpenMP’s Target Construct on GPUs – Exploring Compiler Optimizations

Neural Code Comprehension: A Learnable Representation of Code Semantics

OpenCL Vector Swizzling Optimization under Global Value Numbering

Recent source codes

MSCCL++: A GPU-driven communication stack for scalable AI applications

Benchmark compute shader of Unity against InteropUnityCUDA

Data-efficient LLM Fine-tuning for Code Generation

DeepSpeed is a deep learning optimization library that makes distributed training and inference easy, efficient, and effective.

Large Language Model Powered C-to-CUDA Code Translation: A Novel Auto-Parallelization Framework

GigaAPI: a user-space API that simplifies multi-GPU programming, bridging the gap between the capabilities of parallel GPU systems and the ability of developers to harness their full potential

Coccinelle: a C code transformation engine using SmPL for matches, refactorings, and bug fixing

DuoReduce: MLIR's benchmark

Shamrock: Multi-GPU hydrodynamics for astrophysics

LLMPerf: GPU Performance Modeling meets Large Language Models

Most viewed papers (last 30 days)