Slurm vs Grid Engine
The new FMRIB cluster, Ood, uses the SLURM cluster software and the fsl_sub module now uses the SLURM cluster.
SLURM is significantly different from Grid Engine, in particular, there are no RAM limits for jobs. We STRONGLY recommend that you specify RAM (with fsl_sub's -R option) to ensure efficient use of the cluster, without it, all jobs will default to requesting 15GB of RAM. This also means that the -S/--noramsplit option is meaningless.
To assist with converting scripts that utilise fsl_sub with queue names for the OOD cluster we have provided a script. To enable the script use:
module add queue_migration
And then call with arguments of your original Grid Engine based script and the name of the script to create with time based partition selection.
queue_migration myscript.sh myscript_slurm.sh
If you are targeting the bigmem.q in your script then this will default to requesting 64GB of RAM. If you require more than this then use the --ram option to queue_migration.
Queue mapping
| Jalapeno Queue | Ood Queue |
| veryshort.q | short |
| short.q | short |
| long.q | long |
| verylong.q | long |
| bigmem.q | long (+ memory specifier) |
| interactive.q | Reserved for remote desktop system Can launch interactive tasks on any of the normal queues |
| gpu.q-----------------> | gpu_short |
| |_________________> | gpu_long |
Multi-Threaded Tasks
fsl_sub's native options remain the same, but of note, SLURM does not support parallel environments, so when requesting multi-thread jobs slots you can use -s <number>. If you provide a parallel environment name this will be discarded, so existing scripts should continue to work as is.
Interactive GUI apps
Interactive tasks should be run via the new Open OnDemand virtual desktop facility.