We have existing SLURM documentation here https://rapids.ai/hpc. We should migrate this to the deployment documentation here.
Create a new documentation page called source/hpc/slurm.md with instructions on launching a Dask cluster with GPUs on a SLURM scheduler with dask-jobqueue and/or dask-mpi.
We have existing SLURM documentation here https://rapids.ai/hpc. We should migrate this to the deployment documentation here.
Create a new documentation page called
source/hpc/slurm.md
with instructions on launching a Dask cluster with GPUs on a SLURM scheduler withdask-jobqueue
and/ordask-mpi
.