CUDA
CUDA (formerly Compute Unified Device Architecture) is a parallel computing platform and programming model created by NVIDIA and implemented by the graphics processing units (GPUs) that they produce. CUDA gives developers access to the virtual instruction set and memory of the parallel computational elements in CUDA GPUs.
https://developer.nvidia.com/cuda-toolkit
Available modules
The overview below shows which CUDA installations are available per target architecture in the HPCC module system, ordered based on software version (new to old).
To start using CUDA, load one of these modules using a module load command like:
1 | |
(This data was automatically generated on Thu, 14 Nov 2024 at 12:00:29 EST)
| neoverse_v1 | generic | zen2 | zen3 | zen4 | haswell | skylake_avx512 | |
|---|---|---|---|---|---|---|---|
| Grace Nodes | everywhere (except Grace nodes) | amd20 | amd22 | amd24 | intel16 | intel18,amd20-v100,amd21,intel21 | |
| CUDA/12.4.0 | - | x | - | - | - | - | - |
| CUDA/12.3.0 | - | x | - | - | - | - | - |
| CUDA/12.1.1 | - | x | - | - | - | - | x |
| CUDA/11.7.0 | - | x | - | - | - | - | - |