Blender on Biowulf

Blender is the free and open source 3D creation suite. It supports the entirety of the 3D pipeline—modeling, rigging, animation, simulation, rendering, compositing and motion tracking, even video editing and game creation.

Blender on Biowulf is meant for command-line rendering. The .blend file should be created outside of Biowulf, as none of the nodes have proper graphic cards with support for OpenGL, required for running the GUI.

Documentation
Important Notes

NOTE: Blender running in commandline/batch mode with --background may still attempt to connect graphics to an X11 server. To avoid errors, a "fake" X11 server can be started to intercept these connections:

Xvfb -shmem -screen 0 1280x1024x24 &
export DISPLAY=":0"
Interactive job
Interactive jobs should be used for debugging, graphics, or applications that cannot be run as batch jobs.

Allocate an interactive session and run the program.
Sample session (user input in bold):

[user@biowulf]$ sinteractive --cpus-per-task=8 --gres=gpu:p100:1
salloc.exe: Pending job allocation 46116226
salloc.exe: job 46116226 queued and waiting for resources
salloc.exe: job 46116226 has been allocated resources
salloc.exe: Granted job allocation 46116226
salloc.exe: Waiting for resource configuration
salloc.exe: Nodes cn3144 are ready for job

[user@cn3144 ~]$ Xvfb -shmem -screen 0 1280x1024x24 &
[user@cn3144 ~]$ export DISPLAY=":0"
[user@cn3144 ~]$ module load blender CUDA/12.1
[user@cn3144 ~]$ cp $BLENDER_EXAMPLES/fishy_cat.blend .
[user@cn3144 ~]$ blender -t ${SLURM_CPUS_ON_NODE} -noaudio --background --render-output run/output --render-frame 1 -E CYCLES fishy_cat.blend -- --cycles-device CUDA
...
[user@cn3144 ~]$ exit
salloc.exe: Relinquishing job allocation 46116226
[user@biowulf ~]$

Batch job
Most jobs should be run as batch jobs.

Create a batch input file (e.g. blender.sh). For example:

#!/bin/bash
Xvfb -shmem -screen 0 1280x1024x24 &
export DISPLAY=":0"
module load blender CUDA/12.1
blender -t ${SLURM_CPUS_ON_NODE} -noaudio --background file.blend --render-output run/output -E CYCLES [.blend file] -- --cycles-device CUDA

Submit this job using the Slurm sbatch command.

sbatch -p gpu --gres=gpu:p100:1 [--cpus-per-task=#] [--mem=#] blender.sh