Best practices to go from 1000s of netcdf files to analyses on a HPC cluster?

HPC
32

Actually, I realized it can be a lot simpler. I think you can just take the min of read_chunks and target_chunks. Testing this now.

1 Like