-
Notifications
You must be signed in to change notification settings - Fork 68
Home
Matt Norman edited this page Dec 13, 2021
·
15 revisions
A mini app simulating weather-like flows for training in parallelizing accelerated HPC architectures.
- MPI (C, Fortran, and C++)
- OpenACC Offload (C and Fortran)
- OpenMP Threading (C and Fortran)
- OpenMP Offload (C and Fortran)
- C++ Portability
- CUDA-like approach
- https://github.com/mrnorman/YAKL/wiki/CPlusPlus-Performance-Portability-For-OpenACC-and-OpenMP-Folks
- C++ code works on CPU, Nvidia GPUs (CUDA), and AMD GPUs (HIP)
- Author: Matt Norman (ORNL)
- Isaac Lyngaas (ORNL)