ESMCI / cmeps-cime

This is a "fork" of the cime repository that has the development version of the nuopc CMEPS driver and mediator
Other
0 stars 0 forks source link

Jedwards/addgptl #49

Closed jedwards4b closed 6 years ago

jedwards4b commented 6 years ago

Adds some timers to nuopc driver and adds a smart compare tool for runconfig files.

Test suite: by hand Test baseline: Test namelist changes: Test status: bit for bit

Fixes

User interface changes?:

Update gh-pages html (Y/N)?:

Code review:

rsdunlapiv commented 6 years ago

@jedwards4b How do I turn on the timers and see the output?

jedwards4b commented 6 years ago

Timers are on by default and output goes into the timing subdirectory of the case. However because I still have some key timers missing the timing summary file is still mostly empty. Compare to an mct case and notice that until now the nuopc didn't create a timing directory at all.

rsdunlapiv commented 6 years ago
dunlap@cheyenne1:~/UFSCOMP.oct17/cime/scripts> qcmd -l walltime=04:00:00 -- ./create_test --xml-machine cheyenne --xml-testlist testlist_cmeps.xml --baseline-root /glade/scratch/dunlap/BASELINES --compare oct5
rsdunlapiv commented 6 years ago

@jedwards4b I got a bunch of baseline comparison failures against the oct5 baseline. I doubt this is related to the timers so I must have something wrong. Would you expect to need new baselines after the recent merge of CIME master back into nuopc-cmeps?

jedwards4b commented 6 years ago

Yes, I was just looking into this myself - some data model input files have changed. I think that this is an expected difference and we need to create new baselines. Here is the responsible PR https://github.com/ESMCI/cime/pull/2828

rsdunlapiv commented 6 years ago

Makes sense to me since the compsets with active CAM matched the baselines - so indeed does seem to be related to DATM input files. I'll merge in this PR and then regenerate baselines.

rsdunlapiv commented 6 years ago

Also I did confirm creation of the timing directory with the NUOPC cases.