high performance computing on graphics processing units: hgpu.org

hgpu.org » CUDA

Scaling On-Device GPU Inference for Large Generative Models

Jiuqiang Tang, Raman Sarokin, Ekaterina Ignasheva, Grant Jensen, Lin Chen, Juhyun Lee, Andrei Kulik, Matthias Grundmann

View

Download (PDF)

Tags: AI, Computer science, CUDA, LLM, Machine learning, nVidia, nVidia GeForce RTX 4090, OpenCL

May 4, 2025 by hgpu

Mìmir: A real-time interactive visualization library for CUDA programs

Francisco Carter, Nancy Hitschfeld, Cristóbal A. Navarro

View

Download (PDF)

Tags: Computer science, CUDA, nVidia, nVidia GeForce RTX 2070, Rendering, Visualization, Vulkan

May 4, 2025 by hgpu

Dynamic Memory Management on GPUs with SYCL

Russell K. Standish

View

Download (PDF)

Source codes

Tags: Computer science, CUDA, HIP, Memory model, nVidia, Package, Performance, SYCL

May 4, 2025 by hgpu

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

Aashaka Shah, Abhinav Jangda, Binyang Li, Caio Rocha, Changho Hwang, Jithin Jose, Madan Musuvathi, Olli Saarikivi, Peng Cheng, Qinghua Zhou, Roshan Dathathri, Saeed Maleki, Ziyue Yang

View

Download (PDF)

Source codes

Tags: AI, AMD Radeon Instinct MI300X, ATI, Computer science, CUDA, Heterogeneous systems, HIP, nVidia, nVidia A100, nVidia H100, Package

April 27, 2025 by hgpu

InteropUnityCUDA: A Tool for Interoperability Between Unity and CUDA

David Algis, Berenger Bramas, Emmanuelle Darles, Lilian Aveneau

View

Download (PDF)

Source codes

Tags: Benchmarking, Computer science, CUDA, DirectX, nVidia, nVidia GeForce RTX 2070, OpenGL, Package, Rendering

April 27, 2025 by hgpu

Data-efficient LLM Fine-tuning for Code Generation

Weijie Lv, Xuan Xia, Sheng-Jun Huang

View

Download (PDF)

Source codes

Tags: Code generation, Computer science, CUDA, LLM, nVidia, nVidia A100, Package, Python, PyTorch

April 27, 2025 by hgpu

LithOS: An Operating System for Efficient Machine Learning on GPUs

Patrick H. Coppock, Brian Zhang, Eliot H. Solomon, Vasilis Kypriotis, Leon Yang, Bikash Sharma, Dan Schatzberg, Todd C. Mowry, Dimitrios Skarlatos

View

Download (PDF)

Tags: Computer science, CUDA, Machine learning, nVidia, nVidia A100, nVidia H100, Operating systems

April 27, 2025 by hgpu

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

Masahiro Tanaka, Du Li, Umesh Chand, Ali Zafar, Haiying Shen, Olatunji Ruwase

View

Download (PDF)

Source codes

Tags: Computer science, CUDA, Deep learning, Distributed computing, nVidia, nVidia H100, Package, Prefetch

April 27, 2025 by hgpu

Scalability Evaluation of HPC Multi-GPU Training for ECG-based LLMs

Dimitar Mileski, Nikola Petrovski, Marjan Gusev

View

Download (PDF)

Tags: Computer science, CUDA, Deep learning, HPC, LLM, nVidia, nVidia A100, PTX, PyTorch

April 13, 2025 by hgpu

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

Meng Wang, Jinshuo Liu, Zimeng Xu, Chang Ding, Xiaomeng Huang

View

Download (PDF)

Source codes

Tags: Code generation, Computer science, CUDA, LLM, nVidia, nVidia GeForce RTX 4090, Package

April 13, 2025 by hgpu

GigaAPI for GPU Parallelization

M. Suvarna, O. Tehrani

View

Download (PDF)

Source codes

Tags: Computer science, CUDA, Image processing, nVidia, nVidia RTX 6000 Ada, Package

April 13, 2025 by hgpu

Advances in Semantic Patching for HPC-oriented Refactorings with Coccinelle

Michele Martone, Julia Lawall

View

Download (PDF)

Source codes

Tags: Computer science, CUDA, HIP, HPC, nVidia, OpenMP, Package, Programming Languages

March 30, 2025 by hgpu

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

DuoReduce: MLIR's benchmark

Hardware-Assisted Software Testing and Debugging for Heterogeneous Computing

See all packages

* * *

high performance computing on graphics processing units: hgpu.org

Scaling On-Device GPU Inference for Large Generative Models

Mìmir: A real-time interactive visualization library for CUDA programs

Dynamic Memory Management on GPUs with SYCL

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

InteropUnityCUDA: A Tool for Interoperability Between Unity and CUDA

Data-efficient LLM Fine-tuning for Code Generation

LithOS: An Operating System for Efficient Machine Learning on GPUs

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

Scalability Evaluation of HPC Multi-GPU Training for ECG-based LLMs

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

GigaAPI for GPU Parallelization

Advances in Semantic Patching for HPC-oriented Refactorings with Coccinelle

Recent source codes

PELSI: Power-Efficient Layer-Switched Inference

Ouroboros: Virtualized Queues for dynamic memory management

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

Most viewed papers (last 30 days)