Dear FSL experts,

Hi. We want to do parallel processing with probtrackx2 for global probabilistic tractography. We have a department cluster with multiple nodes and cores but not GPU or SGE. When trying to run probtrackx2 in cluster, our department computer engineer told us that the command cannot use multiple nodes. Furthermore, for single node running or serial job, the command can only control single core. That makes the executing time very long. Therefore, we want to ask for help.

Could you please guide us how to make use of multiple nodes and cores to run FSL commands (e.g. probtrackx2) in cluster? I found a command "fsl_sub", but don't know whether it is what I need. Or do you have other suggestions? Any feedback would be very appreciated. Thanks!

Best regards,

Xinyang