[Ncep.list.nems.announce] NEMS/NMM-B commit

John Michalakes john.michalakes at noaa.gov
Mon Oct 6 17:09:59 UTC 2014


These changes have been committed to the top of the NEMS trunk as r46389.
-John

-----Original Message-----
From: John Michalakes [mailto:john.michalakes at noaa.gov] 
Sent: Monday, October 6, 2014 3:43 AM
To: 'Tom Black'; 'NEMS Developers'
Subject: RE: [Ncep.list.nems.announce] NEMS/NMM-B commit

Proceeding.  Thanks, Tom.  

My proposed commits are:

	1.  Fix to shortwave RRTMG to allow running with TRAPS compiled in.
Fixes 
	    floating point exceptions reported by Dusan and Lin. 
	2.  Fix an OpenMP threading error that was introduced in r45178 in 
	    module_DYNAMICS_ROUTINES.F90 .  Fix is to revert two ending DO
loop 
         ranges back to and ite_h2 and jte_h2.

Affected files are src/atmos/phys/grrad_nmmb.F and
src/atmos/phys/radsw_main_nmmb.F and
src/atmos/nmm/module_DYNAMICS_ROUTINES.F90 .   

Full set of differences relative to top of trunk (r46236) is in the attached
.txt file.  The code is on Tide in /gpfs/tp1/ptmp/John.Michalakes/trunk .

-John

-----Original Message-----
From: ncep.list.nems.announce-bounces at lstsrv.ncep.noaa.gov
[mailto:ncep.list.nems.announce-bounces at lstsrv.ncep.noaa.gov] On Behalf Of
Tom Black
Sent: Sunday, October 5, 2014 11:29 AM
To: NEMS Developers
Subject: Re: [Ncep.list.nems.announce] NEMS/NMM-B commit

Unfortunately there is a failure (non bit-identical answers) in the 
NMM-B's moving nest restart regression test when running with this set 
of changes.  It's probably trivial but I'm not sure how long it will 
take to find. John, please proceed with your commit then I'll get back 
in line when this moving nest restart issue is resolved.

Tom

On 10/3/2014 10:55 AM, Tom Black wrote:
> I would like to commit some small changes to the NMM-B associated with 
> NEMS Trac ticket #111.
>
> (1) A fix for an operational failure in the NAM a few weeks ago. The 
> problem arose with a variable in RRTM going to zero so a check is 
> inserted into the RRTM code to keep that variable's magnitude greater 
> than a certain limit.
>
> (2) Moving parents need to fill in values of child-resolution 
> topography on certain edge V-points of the parent domain.
>
> (3) Modify the change Dusan just committed today to use 2 rather than 
> 8 tasks for the parallel compile.
>
> Answers will change for nearly all NMM-B regression tests.  The code 
> can be found at:
>
> tide: /gpfs/td1/emc/meso/noscrub/Tom.Black/trunk_update
> zeus: /scratch2/portfolios/NCEPDEV/meso/noscrub/Tom.Black/trunk_update
>
> I will plan to commit early next week.
>
> M            46308   tests/rt.sh
> M            46308   src/atmos/phys/radsw_main_nmmb.f
> M            46308   src/atmos/nmm/module_NESTING.F90
>
> Tom
>

_______________________________________________
Ncep.list.nems.announce mailing list
Ncep.list.nems.announce at lstsrv.ncep.noaa.gov
https://lstsrv.ncep.noaa.gov/mailman/listinfo/ncep.list.nems.announce



More information about the Ncep.list.nems.announce mailing list