Scheduled Downtime
On Tuesday 24 October 2023 @ 5pm MT the forums will be in read only mode in preparation for the downtime. On Wednesday 25 October 2023 @ 5am MT, this website will be down for maintenance and expected to return online later in the morning.
Normal Operations
The forums are back online with normal operations. If you notice any issues or errors related to the forums, please reach out to help@ucar.edu

CCSM3 bluefire scripts

Has anyone created their own scripts for compiling and running the code for bluefire? Just downloaded the tarball and I see it contains a batch.ibm.bluesky and run.ibm.bluesky, but there are no bluefire files. And they would be totally different since bluefire runs LSF instead of loadleveler.

Just hate to reinvent the wheel...

Thanks!

Mark Branson
Research Associate
Dept. of Atmospheric Science
Colorado State University
 
Sorry for the late reply, but I haven't checked this Forum lately. Probably you've gone ahead and figured it out on your own.

I had posted some comments on my conversion from Bluevista to Blueice in this thread:

http://bb.cgd.ucar.edu/showthread.php?t=975

Converting from Blueice to Bluefire merely required changing "blueice" to "bluefire" in the appropriate places, excepting for commenting out the environment variable "MP_MSG_API" in your Bluefire runscript.
 
Thanks for your reply. You're right -- I eventually sorta stumbled into the files that I needed. First I found the CCSM wiki page and that eventually led me to the /fis/cgd/cseg/csm/collections area on bluefire itself, and under there are various versions of the CCSM, of which more than a few have the necessary bluefire files. They seem to work fine for me.

I do have another question, though: What are the necessary steps to change namelist input for a continuation run? After running a successful 5-day test, I was wanted to change a few CAM namelist variables and continue the run for another 5 days. Apparently just editing the cam/atm.stdin file doesn't do the trick, so I tried sticking the changes into Buildnml_Prestage/cam.buildnml_prestage.csh and rebuilding, and while that does in fact put the changes into atm.stdin, the changes did not manifest themselves in the continuation 5-day simulation. Do I have to rerun configure as well? It seems counterintuitive to me to have to do any of these steps since one of the main reasons you use a namelist in the first place is to have the ability to make certain changes without having to recompile the source code.

So I must be missing something, eh? ;)

Thanks,
Mark
 
Dear All,

is there somone around, who figured out how to compile, build, test and run CCSM3 on an IBM Power6? I am running in trouble to compile the coupler - having problems to locate the chdir lib.

I am using the beta14 subversion of CCSM3 and I am using the mpxlf90 compiler version 12.1.


Code:
cpp -P -q64 -DAIX -DMSS -DTHREADED_PTHREADS -I. -I/u/b304015/exe/case.16/SourceMods/src.cpl -I/u/b304015/ccsm3_0_1_beta14/models/cpl/cpl6 -I/u/b304015/ccsm3_0_1_
beta14/models/csm_share -I/u/b304015/ccsm3_0_1_beta14/models/csm_share/shr -I/u/b304015/ccsm3_0_1_beta14/models/csm_share/cpl -I.  -I/u/m214003/local/include/ -I
/u/b304015/exe/case.16/lib/include -I/usr/lpp/ppe.poe/include/thread  /u/b304015/ccsm3_0_1_beta14/models/csm_share/shr/shr_sys_mod.F90 > shr_sys_mod.f90
mpxlf90_r  -c -qrealsize=8 -qdpc=e -g -qfullpath  -q64 -O2 -qstrict -qextname -qmaxmem=-1 -qarch=auto -Q -qsigtrap=xl__trcedump -qsuffix=f=f90 -qfree=f90 -I. -I/
u/b304015/exe/case.16/SourceMods/src.cpl -I/u/b304015/ccsm3_0_1_beta14/models/cpl/cpl6 -I/u/b304015/ccsm3_0_1_beta14/models/csm_share -I/u/b304015/ccsm3_0_1_beta
14/models/csm_share/shr -I/u/b304015/ccsm3_0_1_beta14/models/csm_share/cpl -I.  -I/u/m214003/local/include/ -I/u/b304015/exe/case.16/lib/include -I/usr/lpp/ppe.p
oe/include/thread  shr_sys_mod.f90  
"shr_sys_mod.f90", 1516-036 (S) Entity chdir_ has undefined type.
** shr_sys_mod   === End of Compilation 1 ===


Is there someone around, having a running version on Power6 - what have you changed?

Thanks in advance for your help,
Tim
 
Top