Claude/3d export vertical chunking 9jb nb#221
Closed
tomsail wants to merge 84 commits into
Closed
Conversation
Turned this option on as default from now on
* CF: out2d validates for ics=1 and iof_ugrid=2 * 3D outputs working for iof_ugrid=0 * Tested for all iof_ugrid = (/1,2,3/) and ics=1 * New iof_ugrid tested with ics=2 and optional metadata.nc input * Documented iof_ugrid=2 and metadata.nc * MAke sure that CRS is transferred from metadata.nc even for ics=1 (when it does not exist in original output) * Simplified to have only iof_ugrid = 0, 1 * Added full reference, make empty institution and license by default * Cleaned up CF output on iof_ugrid=0 * Remove redundant createion_date from variable attributes * Keep iof_ugrid=0 with minimal metadata, keep iof_ugrid=1 with complete data+metadata, new iof_ugrid=2 with complete metadata and referenced data in 3D * Reverted renameing of add_mesh_attributes
Added a link to the live manual for SCHISM in ESMF documentation.
) These variables are incorrectly declared as integers but are used as real values (e.g., veg_cw=1.5). When users specify float values in param.nml, Fortran fails to parse the namelist: Error termination. Backtrace: At line 537 of file .../src/Hydro/schism_init.F90 (unit = 15, file = './/param.nml') Fortran runtime error: Cannot match namelist object name Move veg_lai and veg_cw from integer to real(rkind) declaration block.
in ref_ts_scale.gr3
* added variable for CICE coupling * fixed imp. * updated CICE comments * minor fix for marsh * remove duplicate code --------- Co-authored-by: Joseph Smith <smithj28@oregonstate.edu>
…ripts (STOFS3D-Atl-v3.1 or earlier).
…FS-3D versions (v3.1 and beyond), while maintaining compatibility with previous versions. The previous STOFS-3D-Atl-shadow-VIMS is now called STOFS-3D-Atl-Setup, to support not only shadow forecast but other applications.
3D scribed outputs (temperature, salinity, velocity, tracers, ...) were written with a single chunk per record covering the whole volume (nvrt x nhoriz x 1). On large meshes this hits the 4 GB chunk ceiling and makes level-wise reads with xarray / ParaView load the entire volume just to slice one layer. Add an SCHOUT namelist switch `nchunk_vrt` controlling vertical chunking in nc_writeout3D: 0 - legacy behaviour (whole-volume chunk per time step) N>0 - N layers per chunk; default 1 (one layer per chunk) Also auto-shrink the horizontal chunk dimension when the configured shape exceeds the 4 GB netCDF-4 limit, instead of aborting outright.
Member
|
Thx @tomsail. It shows changes in ~800 files. Please submit PR only for the few files/changes you made to the chunking. Thx |
Author
|
Yes sorry for this.. I forgot to sync my fork. Hence why I closed this PR immediately |
Member
|
I saw your new PR; thx Thomas! |
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
13 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
No description provided.