search

hubmapconsortium / codex-pipeline

CODEX data processing code
GNU General Public License v3.0
11 stars 4 forks source link

Increase Slurm time limit for large datasets #3

Closed mkeays closed 3 years ago

mkeays commented 4 years ago

Larger datasets (e.g. those with multiple regions) require processing times that exceed the current maximum of 4 hours. We could default to a longer time, or somehow calculate a time limit based on the amount of data to process, which might be better if we end up with different GPU queues in future.

mkeays commented 4 years ago

Not sure whether this is still needed since we switched to CWL.