MITgcm Setup: Difference between revisions

From Fluids Wiki
Jump to navigation Jump to search
m (Tghill moved page MITgcmSetup to MITgcm Setup: Rename page to be more descriptive)
No edit summary
 
(3 intermediate revisions by the same user not shown)
Line 1: Line 1:
This page is intended to help someone get started with MITgcm for the first time.
This page will help you get started with MITgcm for the first time.
 
== Acquiring the code ==
The MITgcm is hosted on [https://github.com/MITgcm/MITgcm GitHub]. I highly recommend cloning the code and updating your model code as the main repository is updated. The model is still being actively improved. For example, in just 4 months:
 
* A bug in the MITgcmutils python package was fixed to no longer invoke a FutureWarning
* The description of the SIZE.h file was significantly improved
* The defaults of the SEAICE package are in the process of being fixed (active Pull Request)
 
To get started, log into the machine you want to set up the code on (ie, Graham, Hood, one of the Mountain Lake machines). Change directories onto a project disk where you want to put the code (If you are on Graham this must done from a login node, since the compute nodes don't have internet access). Run
 
    git clone https://github.com/MITgcm/MITgcm.git
 
Once this completes, you should have a directory <code>MITgcm</code> which contains
 
    MITgcm/
   
        LICENSE.txt
        README.md
        doc/
        eesupp/
        jobs/
        lsopt/
        model/
        optim/
        pkg/
        tools/
        utils/
        verification/
 
Now you're ready to run the [[MITgcm_Tutorials|tutorials]] and start designing and running your own cases.
 
The [https://git.uwaterloo.ca/tghill/gcmpy gcmpy package] has some useful tools for visualizing and handling MITgcm data. Its documentation lives on [https://gcmpy.readthedocs.io/en/latest/ Read the Docs].

Latest revision as of 11:08, 16 August 2018

This page will help you get started with MITgcm for the first time.

Acquiring the code

The MITgcm is hosted on GitHub. I highly recommend cloning the code and updating your model code as the main repository is updated. The model is still being actively improved. For example, in just 4 months:

  • A bug in the MITgcmutils python package was fixed to no longer invoke a FutureWarning
  • The description of the SIZE.h file was significantly improved
  • The defaults of the SEAICE package are in the process of being fixed (active Pull Request)

To get started, log into the machine you want to set up the code on (ie, Graham, Hood, one of the Mountain Lake machines). Change directories onto a project disk where you want to put the code (If you are on Graham this must done from a login node, since the compute nodes don't have internet access). Run

   git clone https://github.com/MITgcm/MITgcm.git

Once this completes, you should have a directory MITgcm which contains

   MITgcm/
   
       LICENSE.txt
       README.md
       doc/
       eesupp/
       jobs/
       lsopt/
       model/
       optim/
       pkg/
       tools/
       utils/
       verification/

Now you're ready to run the tutorials and start designing and running your own cases.

The gcmpy package has some useful tools for visualizing and handling MITgcm data. Its documentation lives on Read the Docs.