Hi all, my question is a follow up on the solution posted here:
In summary I am in a similar position of using parallel pools on a HPC cluster. I am able to use parallel pools after deleting the \.matlab folder in my home directory. This is rather limiting however, as I need to run many jobs often, and concurrently. This would require me to clear the directory every time a new job is run, otherwise parallel pools will not start. Is there a way to automate this process? or perhaps not need it in the first place? If I cleared/deleted \.matlab while a another parallel pool job is running, will it be affected?
Thanks in advance for your suggestions.
I'm not very skilled in PBS/Slurm commands so forgive me if there's an easy way to do this in the job script.
Key phrase for future: "failed to locate and destroy old interactive jobs"