n-jobs in generic-prep function

[TheUwe]

While pre-processing resting state fMRI data, I tried to limit the processing done in parallel, since my cluster allows only for a limited amount of capacity used. Therefore, I wanted to use the implemented parameter '--n-jobs' of the 'generic-prep' function. Unfortunately, I couldn't manage to let the parameter be accepted.

I tried my basic code SAMRI generic-prep -n-jobs 4 -o Exp1 -f '{"acquisition":["rsfMRI"],"subject":["169"]}' -s '{"acquisition":["FLASH"],"subject":["169"]}' Exp1/bids samriTemplates/dsurqec_200micron.nii with various other options of -n-jobs that all led to the following error: SAMRI: error: unrecognized arguments: -n-jobs samriTemplates/dsurqec_200micron.nii

Running this code without the n-jobs parameter leads to an error message:

Message from root@eu-login-10-ng on pts/2 at 10:40 ... You are overloading this login node (2 threads N4BiasFieldCorr + 9 threads SAMRI + 1 thread python on 4 CPUS). Please DO NOT use more than 4 threads + processes on login nodes ! Please use LSF, if you need more resources.

[TheChymera]

Thank you for the report. I have reviewed the code and it should work. Please specify:

• The exact command which you used (the command you posted above does not specify a jobs number)
• How you know that this used “too many cores” --- ideally while specifying the number of cores.

[TheChymera]

@TheUwe is this still relevant?