add esmf810 updates

[junwang-noaa]

Add GFSv16 nems updates for performance tuning

Associated PRs:

fv3atm PR#160 ufs-weather-model PR #188

this NEMS #PR76 needs to be committed before ufs-weather-model PR #188.

[junwang-noaa]

The change is not to make ww3 run faster, without it the ufs-weather can't even compile with ww3 using gnu make. Will any person from WW3 take a look at it? We can use cmake build though.

On Sat, Aug 22, 2020 at 6:30 PM Jessica Meixner notifications@github.com wrote:

@JessicaMeixner-NOAA commented on this pull request.

In src/incmake/component_WW3.mk https://github.com/NOAA-EMC/NEMS/pull/76#discussion_r475144797:

@@ -37,7 +37,7 @@ WW3_ALL_OPTS= \ $(ww3_mk): configure +$(MODULE_LOGIC) ; set -x ; cd $(WW3_SRCDIR)/esmf ; \ export $(WW3_ALL_OPTS) ; \

• exec $(MAKE) -j 1 WW3_COMP="$(WW3_CONFOPT)" ww3_nems
• exec $(MAKE) -j 1 WW3_COMP="$(WW3_CONFOPT)" ww3_nemslibonly

While this change will make WW3 run faster, there currently is not another work around for building the other WW3 jobs in the workflow. Until this is there, I don't think this is a good idea.

— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub https://github.com/NOAA-EMC/NEMS/pull/76#pullrequestreview-472966192, or unsubscribe https://github.com/notifications/unsubscribe-auth/AI7D6TMMSGCCFVNC3KIYAZLSCBBIDANCNFSM4QFRZLOQ .

[SMoorthi-emc]

Jun, Is this compilation not using “compile.sh”? I wonder why I can compile without this change under s2s. Moorthi

Sent from my iPhone

On Aug 22, 2020, at 9:39 PM, Jun Wang notifications@github.com wrote:

The change is not to make ww3 run faster, without it the ufs-weather can't even compile with ww3 using gnu make. Will any person from WW3 take a look at it? We can use cmake build though.

On Sat, Aug 22, 2020 at 6:30 PM Jessica Meixner notifications@github.com wrote:

@JessicaMeixner-NOAA commented on this pull request.

In src/incmake/component_WW3.mk https://github.com/NOAA-EMC/NEMS/pull/76#discussion_r475144797:

@@ -37,7 +37,7 @@ WW3_ALL_OPTS= \ $(ww3_mk): configure +$(MODULE_LOGIC) ; set -x ; cd $(WW3_SRCDIR)/esmf ; \ export $(WW3_ALL_OPTS) ; \

• exec $(MAKE) -j 1 WW3_COMP="$(WW3_CONFOPT)" ww3_nems
• exec $(MAKE) -j 1 WW3_COMP="$(WW3_CONFOPT)" ww3_nemslibonly

While this change will make WW3 run faster, there currently is not another work around for building the other WW3 jobs in the workflow. Until this is there, I don't think this is a good idea.

— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub https://github.com/NOAA-EMC/NEMS/pull/76#pullrequestreview-472966192, or unsubscribe https://github.com/notifications/unsubscribe-auth/AI7D6TMMSGCCFVNC3KIYAZLSCBBIDANCNFSM4QFRZLOQ .

— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub, or unsubscribe.

[junwang-noaa]

Moorthi, the new NCEPLIBS build is only in ufs-weather and not used by s2s yet. But we'd like to use the same NCEPLIBS in both models in the future. The issue in current ufs-weather model is that ww3 needs to build both ww3 lib required for the model and the ww3 pre- and post- procoessing executable, and there is some issue with these executable to use the new NCEPLIBS.

On Sun, Aug 23, 2020 at 7:44 AM SMoorthi-emc notifications@github.com wrote:

Jun, Is this compilation not using “compile.sh”? I wonder why I can compile without this change under s2s. Moorthi

Sent from my iPhone

On Aug 22, 2020, at 9:39 PM, Jun Wang notifications@github.com wrote:

The change is not to make ww3 run faster, without it the ufs-weather can't even compile with ww3 using gnu make. Will any person from WW3 take a look at it? We can use cmake build though.

On Sat, Aug 22, 2020 at 6:30 PM Jessica Meixner < notifications@github.com> wrote:

@JessicaMeixner-NOAA commented on this pull request.

In src/incmake/component_WW3.mk https://github.com/NOAA-EMC/NEMS/pull/76#discussion_r475144797:

@@ -37,7 +37,7 @@ WW3_ALL_OPTS= \ $(ww3_mk): configure +$(MODULE_LOGIC) ; set -x ; cd $(WW3_SRCDIR)/esmf ; \ export $(WW3_ALL_OPTS) ; \

• exec $(MAKE) -j 1 WW3_COMP="$(WW3_CONFOPT)" ww3_nems
• exec $(MAKE) -j 1 WW3_COMP="$(WW3_CONFOPT)" ww3_nemslibonly

While this change will make WW3 run faster, there currently is not another work around for building the other WW3 jobs in the workflow. Until this is there, I don't think this is a good idea.

— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub <https://github.com/NOAA-EMC/NEMS/pull/76#pullrequestreview-472966192 , or unsubscribe < https://github.com/notifications/unsubscribe-auth/AI7D6TMMSGCCFVNC3KIYAZLSCBBIDANCNFSM4QFRZLOQ

.

— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub, or unsubscribe.

— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub https://github.com/NOAA-EMC/NEMS/pull/76#issuecomment-678764006, or unsubscribe https://github.com/notifications/unsubscribe-auth/AI7D6TMRJ7EOT6AL2754VWDSCD6LJANCNFSM4QFRZLOQ .

[junwang-noaa]

There was a discussion with Gerhard on this, I'd like to have a version matching GFS.v16, then delete it in a later commit. I will create an issue in nems so that we won't forget.

On Mon, Aug 24, 2020 at 8:31 AM Dusan Jovic notifications@github.com wrote:

@DusanJovic-NOAA commented on this pull request.

In src/module_EARTH_GRID_COMP.F90 https://github.com/NOAA-EMC/NEMS/pull/76#discussion_r475564323:

@@ -256,10 +256,12 @@ SUBROUTINE EARTH_REGISTER(EARTH_GRID_COMP,RC_REG) specLabel=Driver_label_SetRunClock, specRoutine=NUOPC_NoOp, rc=RC_REG) ESMF_ERR_RETURN(RC,RC_REG)

+#if 0

If this call is not needed anymore, delete it.

— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub https://github.com/NOAA-EMC/NEMS/pull/76#pullrequestreview-473417456, or unsubscribe https://github.com/notifications/unsubscribe-auth/AI7D6TM6JASB4APHJZW55QTSCJMTBANCNFSM4QFRZLOQ .