rapidsai

Link to section 'Description' of 'rapidsai' Description

The RAPIDS suite of software libraries gives you the freedom to execute end-to-end data science and analytics pipelines entirely on GPUs. It relies on NVIDIA® CUDA® primitives for low-level compute optimization, but exposes that GPU parallelism and high-bandwidth memory speed through user-friendly Python interfaces.

Link to section 'Versions' of 'rapidsai' Versions

Scholar: 0.12, 0.13, 0.14, 0.15, 0.16, 0.17, 21.06, 21.10
Gilbreth: 0.12, 0.13, 0.14, 0.15, 0.16, 0.17, 21.06, 21.10
Anvil: 0.12, 0.13, 0.14, 0.15, 0.16, 0.17, 21.06, 21.10

Link to section 'Module' of 'rapidsai' Module

You can load the modules by:

module load ngc
module load rapidsai

Link to section 'Example job' of 'rapidsai' Example job

Using #!/bin/sh -l as shebang in the slurm job script will cause the failure of some biocontainer modules. Please use #!/bin/bash instead.

To run rapidsai on our clusters:

#!/bin/bash
#SBATCH -A myallocation     # Allocation name
#SBATCH -t 1:00:00
#SBATCH -N 1
#SBATCH -n 1
#SBATCH -c 8
#SBATCH --gpus-per-node=1
#SBATCH --job-name=rapidsai
#SBATCH --mail-type=FAIL,BEGIN,END
#SBATCH --error=%x-%J-%u.err
#SBATCH --output=%x-%J-%u.out

module --force purge
ml ngc rapidsai