Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

restart file IO problems found in RRFS cycling test #2020

Closed
hu5970 opened this issue Nov 30, 2023 · 0 comments · Fixed by NOAA-EMC/fv3atm#726 or #2021
Closed

restart file IO problems found in RRFS cycling test #2020

hu5970 opened this issue Nov 30, 2023 · 0 comments · Fixed by NOAA-EMC/fv3atm#726 or #2021
Labels
bug Something isn't working

Comments

@hu5970
Copy link

hu5970 commented Nov 30, 2023

Description

When test the latest ufs-weather-model (41b1724) code for RRFS application. We found several issues related to the restart file and output file IO:

  1. need more memory for output quilting. This is solved by adding more core for quilting.
  2. the "coupler.res" is not the last file to write in the restart files. This caused some trouble in workflow to decide if all the restart files are available or not. It is better to write the "coupler.res" after all other restart files so it can be used as a flag for complete restart files writing process.
  3. The chunksize of restart files are not right. Several RRFS applications use parallel netcdf IO that depends on right chunksize (nx * ny). This code cannot generate right chunksize for RRFS applications when "quilting_restart=.true.".
  4. calendar type used by RRFS needs to set to 2 ('julian') instead of 3 ('gregorian') when generates "coupler.res".

To Reproduce:

All those issues were found in RRFS_A parallel test on CACTUS.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
bug Something isn't working
Projects
None yet
1 participant