You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Provide a clear and concise description of what the bug is.
A slowness was found when using the recent ufs-weather-model develop branch. It was identified that it is from ufs-weather-mode PR #425, so far it only impacts the threading test, it is not seen in the tests with 1 thread. In LAM parallel run when using this revision, the running time is increase from 4800s->7700s. It is also seen in the ufs-weather-model threading test fv3_ccpp_2threads test.
To Reproduce:
What compilers/machines are you seeing this with?
Check out ufs-weather-model, run fv3_ccpp_2threads test, the run time is increased to 49s compared to 42s from previous revision.
The text was updated successfully, but these errors were encountered:
#474)
* Export OMP settings in the make_orog ex-script and change the default KMP_AFFINITY for the make_orog task to match WCOSS.
* Add export command to all OMP settings in each ex-script.
Description
Provide a clear and concise description of what the bug is.
A slowness was found when using the recent ufs-weather-model develop branch. It was identified that it is from ufs-weather-mode PR #425, so far it only impacts the threading test, it is not seen in the tests with 1 thread. In LAM parallel run when using this revision, the running time is increase from 4800s->7700s. It is also seen in the ufs-weather-model threading test fv3_ccpp_2threads test.
To Reproduce:
What compilers/machines are you seeing this with?
Check out ufs-weather-model, run fv3_ccpp_2threads test, the run time is increased to 49s compared to 42s from previous revision.
The text was updated successfully, but these errors were encountered: