...
@HLRN: Load prerequisite modules
Codeblock module load gcc/9.2.0 module load openmpi/gcc.9 module load anaconda3 module load llvm module load paraview
@HLRN: Start interactive job. If a lot of memory is needed, especially the fat nodes are suitable:
Codeblock srun -N1-partition=large96 --partitionnodes=large961 --ntasks-per-node=96 --mem=747000mb --pty bash
@HLRN: Start Paraview-Server on your fat node bfn####
Codeblock mpirun -n $SLURM_TASKS_PER_NODE pvserver
@local: Start ssh tunnel on your preferred terminal to access fat node via HLRN login node
Codeblock ssh -N -L 11111:bfn####:11111 <user>@blogin.hlrn.de
Leave this terminal window open to keep the tunnel running. Before setting up a ssh tunnel check if your standard ssh login works. If you are a windows user without a proper terminal we recommend MobaXterm.
@local: Start your ParaView client GUI and access your Paraview-Server at
Codeblock localhost:11111
Please note: the version of your local ParaView client has to be the same as the remote Paraview-Server. You can download various ParaView versions here. After calling "module load paraview" the version of the loaded module will be displayed. Currently it is 5.7.
...