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

cam5 build as a standalone mode

Hi all,

I am learning how to build cam5.

What is the difference when we running cam5 as a part of CCSM and as a standalone mode.
If I want to run cam5 as a standalone mode, which steps should to do and how to create new case.
 

eaton

CSEG and Liaisons
Just to be clear, there is no CAM5 in CCSM. CAM5 is only in the CESM.

Running CAM in a "standalone" mode is somewhat ambiguous.

The CAM standalone configuration, i.e., CAM with a data ocean (docn), active land (clm), and a thermodynamic sea ice component (cice), can be run using the CESM scripts, or using CAM standalone scripts. Running with the CESM scripts is the recommended way since these scripts provide a lot of support for doing production runs. To run CAM standalone from the CESM scripts you just specify one of the "F" compsets to the create_newcase command.

CAM standalone scripts (there are some example standalone scripts in models/atm/cam/bld/) require a more expert level of knowledge. They are only capable of running CAM in the standalone configuration, and are most suitable for development work rather than production runs.

The only difference between running the CAM standalone configuration from the CESM scripts versus from the CAM standalone scripts is the scripts themselves. The driver code and all components are identical regardless of the scripts used to build the model.
 
But when I change the compset to 'F' and try to run a standalone cam5, it still need to build the mct and pio library. It seemed have the same mistake as my cesm building process.
gfortran: big_endian: No such file or directory
gfortran: byterecl: No such file or directory
gfortran: unrecognized option '-r8'
gfortran: unrecognized option '-convert'
cc1: error: unrecognized command line option "-ftz"
gmake[1]: *** [pio_kinds.o] Error 1
gmake[1]: Leaving directory `/disk4/una/output/test/pio'
gmake: *** [all] Error 2
 

eaton

CSEG and Liaisons
My guess is that you're specifying the generic_linux_intel machine when you execute the create_newcase command, but that the mpif90 in your path is invoking the gfortran compiler. If you want to use intel compilers you need to make sure that the mpi library was built with those compilers, and that your path is correctly set up to use the mpif90 that was created when mpi was installed using intel compilers. Note that the path can be specified in the scripts/ccsm_utils/Machines/env_machopts.generic_linux_intel file.

You are correct that mct and pio will be built whether you use the cesm scripts or use a cam standalone script.
 
Top