Thomas SchwingeNov 15
All set for Supercomputing 2025 conference! #SC25 @baylibre #GCC #OpenACC #OpenMP #HPC
HGPU groupOct 5

VibeCodeHPC: An Agent-Based Iterative Prompting Auto-Tuner for HPC Code Generation Using LLMs

#CUDA #OpenMP #OpenACC #HPC #LLM #CodeGeneration #Package

https://hgpu.org/?p=30280

VibeCodeHPC: An Agent-Based Iterative Prompting Auto-Tuner for HPC Code Generation Using LLMs

We propose VibeCodeHPC, an automatic tuning system for HPC programs based on multi-agent LLMs for code generation. VibeCodeHPC tunes programs through multi-agent role allocation and iterative promp…

hgpu.org
HGPU groupSep 7

GPU-acceleration of the Discontinuous Galerkin Shallow Water Equations Solver (DG-SWEM) using CUDA and OpenACC

#CUDA #OpenACC #CFD #FluidDynamics #Package

https://hgpu.org/?p=30169

GPU-acceleration of the Discontinuous Galerkin Shallow Water Equations Solver (DG-SWEM) using CUDA and OpenACC

This paper presents a porting of DG-SWEM, a discontinuous Galerkin solver for coastal ocean circulation, and in particular storm surge, to GPU using two separate approaches: CUDA Fortran and OpenAC…

hgpu.org
HGPU groupSep 7

AnnotationGym: A Generic Framework for Automatic Source Code Annotation

#HLS #FPGA #OpenMP #OpenACC

https://hgpu.org/?p=30167

AnnotationGym: A Generic Framework for Automatic Source Code Annotation

A common approach to code optimization is to insert compiler hints in the source code using annotations. Two major challenges with using annotations effectively are their complexity and lack of por…

hgpu.org
HGPU groupAug 10

ConTraPh: Contrastive Learning for Parallelization and Performance Optimization

#OpenCL #OpenACC #OpenMP #HPC #LLM #CodeGeneration

https://hgpu.org/?p=30084

ConTraPh: Contrastive Learning for Parallelization and Performance Optimization

With the advancement of HPC platforms, the demand for high-performing applications continues to grow. One effective way to enhance program performance is through parallelization. However, fully lev…

hgpu.org
Leibniz Supercomputing CentreJul 23

Unique opportunity for #AI and #ML engineers and data scientists!

Together with #NVIDIA and #OpenACC organization, the #EuroCC National Competence Centres of Austria, Germany, and Poland will host an Open AI Hackathon for teams who want to accelerate, optimize, and scale their real-world projects leveraging the latest technologies.

👉 EUROCC AI HACKATHON
📆 October 14-23, 2025

Apply until August 5, 2025: https://www.openhackathons.org/s/siteevent/a0CUP000013Tp8f2AC/se000375

HGPU groupDec 24, 2024

Reproducible Study and Performance Analysis of GPU Programming Paradigms: OpenACC vs. CUDA in Key Linear Algebra Computations

#CUDA #OpenACC #HPC #BLAS #Performance #Package

https://hgpu.org/?p=29611

Reproducible Study and Performance Analysis of GPU Programming Paradigms: OpenACC vs. CUDA in Key Linear Algebra Computations

Scientific and engineering problems are frequently governed by partial differential equations; however, the analytical solutions of these equations are often impractical, thereby forcing the adopti…

hgpu.org
HGPU groupDec 8, 2024

Unified schemes for directive-based GPU offloading

#OpenMP #OpenACC #Package

https://hgpu.org/?p=29584

Unified schemes for directive-based GPU offloading

GPU is the dominant accelerator device due to its high performance and energy efficiency. Directive-based GPU offloading using OpenACC or OpenMP target is a convenient way to port existing codes or…

hgpu.org
HGPU groupSep 29, 2024

OpenACC offloading of the MFC compressible multiphase flow solver on AMD and NVIDIA GPUs

#OpenACC #CFD #MPI #Package

https://hgpu.org/?p=29426

OpenACC offloading of the MFC compressible multiphase flow solver on AMD and NVIDIA GPUs

GPUs are the heart of the latest generations of supercomputers. We efficiently accelerate a compressible multiphase flow solver via OpenACC on NVIDIA and AMD Instinct GPUs. Optimization is accompli…

hgpu.org
HGPU groupAug 18, 2024

Portability of Fortran’s ‘do concurrent’ on GPUs

#OpenACC #OpenMP #Fortran #Portability #Package

https://hgpu.org/?p=29345

Portability of Fortran’s ‘do concurrent’ on GPUs

There is a continuing interest in using standard language constructs for accelerated computing in order to avoid (sometimes vendor-specific) external APIs. For Fortran codes, the {tt do concurrent}…

hgpu.org