olcf-tutorials / local_mpi_to_gpu
How to use node-local MPI rank IDs to manually map MPI ranks to GPUs
☆13Updated 4 years ago
Related projects ⓘ
Alternatives and complementary repositories for local_mpi_to_gpu
- Molecular dynamics proxy application based on Kokkos☆31Updated 4 months ago
- QMCPACK miniapp: a simplified real space QMC code for algorithm development, performance portability testing, and computer science experi…☆27Updated 3 months ago
- MiniMD Molecular Dynamics Mini-App☆48Updated 3 months ago
- ☆14Updated this week
- This tutorial demonstrates how to use CUDA-Aware MPI☆37Updated last year
- OpenMP vs Offload☆21Updated last year
- Logger for MPI communication☆26Updated last year
- Benchmark implementation of CosmoFlow in TensorFlow Keras☆20Updated 9 months ago
- Comb is a communication performance benchmarking tool.☆24Updated last year
- Intermediate MPI lesson☆26Updated last year
- Very-Low Overhead Checkpointing System☆54Updated 3 weeks ago
- ☆10Updated 3 months ago
- Training examples for SYCL☆38Updated last week
- CPE change log and release notes☆26Updated 2 months ago
- Distributed View Extension for Kokkos