Closed xylar closed 1 week ago
See https://github.com/E3SM-Ocean-Discussion/E3SM/pull/89 for relevant discussion.
The new e3sm_cryo_developer
suite has been tested on these machines with these test-scheduler total times:
Separately, a successful SMS_D_Ld5.T62_oQU240wLI.GMPAS-DIB-IAF-DISMF.chrysalis_intel
test verified correct reading in of new DIB and DISMF datasets.
I tried to test this on Compy but can't log in (a frequent problem for me) and I don't have time to debug this. Anyone else able to log in and try this out?
I got onto Compy again and am testing this now.
I successfully ran the e3sm_cryo_developer
suite on Compy in about 57 minutes:
test-scheduler took 3433.970710992813 seconds
@proteanplanet, could you review this when you have time?
@proteanplanet -- I looked at the output from the failed run and I think the job is just too large to run on four anvil nodes. I'm testing on chrysalis
It passed on chrysalis using the config_pes_tests.xml layout:
PASS SMS_D_Ld1.TL319_IcoswISC30E3r5.GMPAS-JRA1p5-DIB-PISMF.chrysalis_intel.mpaso-jra_1958 CREATE_NEWCASE
PASS SMS_D_Ld1.TL319_IcoswISC30E3r5.GMPAS-JRA1p5-DIB-PISMF.chrysalis_intel.mpaso-jra_1958 XML
PASS SMS_D_Ld1.TL319_IcoswISC30E3r5.GMPAS-JRA1p5-DIB-PISMF.chrysalis_intel.mpaso-jra_1958 SETUP
PASS SMS_D_Ld1.TL319_IcoswISC30E3r5.GMPAS-JRA1p5-DIB-PISMF.chrysalis_intel.mpaso-jra_1958 SHAREDLIB_BUILD time=134
PASS SMS_D_Ld1.TL319_IcoswISC30E3r5.GMPAS-JRA1p5-DIB-PISMF.chrysalis_intel.mpaso-jra_1958 MODEL_BUILD time=403
PASS SMS_D_Ld1.TL319_IcoswISC30E3r5.GMPAS-JRA1p5-DIB-PISMF.chrysalis_intel.mpaso-jra_1958 SUBMIT
PASS SMS_D_Ld1.TL319_IcoswISC30E3r5.GMPAS-JRA1p5-DIB-PISMF.chrysalis_intel.mpaso-jra_1958 RUN time=732
PASS SMS_D_Ld1.TL319_IcoswISC30E3r5.GMPAS-JRA1p5-DIB-PISMF.chrysalis_intel.mpaso-jra_1958 MEMLEAK insufficient data for memleak test
PASS SMS_D_Ld1.TL319_IcoswISC30E3r5.GMPAS-JRA1p5-DIB-PISMF.chrysalis_intel.mpaso-jra_1958 SHORT_TERM_ARCHIVER
I rebased to address the conflict with e3sm_landice_developer
, which was recently added in the same location.
Passes new e3sm_cryo_developer suite on chrysalis, as well as sanity testing. Merged to next
verified the new files are on the inputdata repo and have the correct permissions
merged to master and new baselines blessed -- except for mappy which may need help downloading the new files
Thank you @jonbob, @cbegeman, @proteanplanet and @darincomeau for helping to make this happen!
This merge adds data iceberg (DIB) and data ice-shelf melt flux (DISMF) files for the oQU240wLI mesh.
It then uses these files in tests added to a new
e3sm_cryo_developer
test suite.