You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Then activate the environment; you'll need to do this each time you launch a new bash session.
87
87
```{bash}
@@ -100,33 +100,35 @@ Successfully installed Isca
100
100
101
101
### Compiling for the first time
102
102
103
-
At Exeter University, Isca is compiled using:
104
-
105
-
* Intel Compiler Suite 14.0
106
-
* OpenMPI 10.0.1
107
-
* NetCDF 4.3.3.1
108
-
* git 2.1.2
103
+
By installing Isca using the `ci/environment-py3.12_frozen.yml` file, you will have installed everything you need to run Isca, including the gfortran compiler, openmpi and netcdf. Whilst this may not be the optimal way of running Isca on your machine, it should work without too much extra effort.
109
104
110
-
Different workstations/servers at different institutions will have different compilers and libraries available. The Isca framework assumes you have something similar to our stack at Exeter, but provides a hook for you to configure the environment in which the model is run.
105
+
Before Isca is compiled/run, an environment is first configured which loads the specific compilers and libraries necessary to build the code. This done by setting the environment variable `GFDL_ENV` in your session. There is an option within Isca to use set `GFDL_ENV=ubuntu_conda`, which is setup to use the gfortran compiler you installed via the environment file.
111
106
112
-
Before Isca is compiled/run, an environment is first configured which loads the specific compilers and libraries necessary to build the code. This done by setting the environment variable `GFDL_ENV` in your session.
113
-
114
-
For example, on the EMPS workstations at Exeter, I have the following in my `~/.bashrc`:
107
+
To make use of this environment, you should add a version of the following to your `~/.bashrc`:
115
108
116
109
```{bash}
117
110
# directory of the Isca source code
118
111
export GFDL_BASE=/scratch/jamesp/Isca
119
-
# "environment" configuration for emps-gv4
120
-
export GFDL_ENV=emps-gv
112
+
# "environment" configuration for use with ubuntu-conda
113
+
export GFDL_ENV=ubuntu_conda
121
114
# temporary working directory used in running the model
122
-
export GFDL_WORK=/scratch/jamesp/gfdl_work
115
+
export GFDL_WORK=/scratch/jamesp/isca_work
123
116
# directory for storing model output
124
-
export GFDL_DATA=/scratch/jamesp/gfdl_data
117
+
export GFDL_DATA=/scratch/jamesp/isca_data
125
118
```
126
119
127
-
The value of `GFDL_ENV` corresponds to a file in `src/extra/env` that is sourced before each run or compilation. For an example that you could adapt to work on your machine, see `src/extra/env/emps-gv`.
120
+
The value of `GFDL_ENV` corresponds to a file in `src/extra/env` that is sourced before each run or compilation.
121
+
122
+
You may wish to configure your own way to run Isca using locally available compilers, e.g. the Intel compilers. An example using such a setup is available here - `src/extra/env/emps-gv`.
123
+
124
+
At Exeter University, Isca is compiled using:
125
+
126
+
* Intel Compiler Suite 14.0
127
+
* OpenMPI 10.0.1
128
+
* NetCDF 4.3.3.1
129
+
* git 2.1.2
128
130
129
-
We are not able to provide support in configuring your environment at other institutions other than Exeter University - we suggest that you contact your friendly local sysops technician for guidance in getting the compilers and libraries collated if you are not sure how to proceed.
131
+
Different workstations/servers at different institutions will have different compilers and libraries available. The Isca framework should run 'out of the box' using the gfortran compilers installed from the environment file. However, if you want to install the model using your own configuration then we are not able to provide support in configuring your environment. We suggest that you contact your friendly local sysops technician for guidance in getting the compilers and libraries collated if you are not sure how to proceed.
130
132
131
133
If you work at another large institution and have successfully compiled and run Isca, we welcome you to commit your own environment config to `/src/extra/env/my-new-env` for future scientists to benefit from and avoid the pain of debugging compilation!
Copy file name to clipboardExpand all lines: docs/source/begginers_guide.rst
+2-2Lines changed: 2 additions & 2 deletions
Display the source diff
Display the rich diff
Original file line number
Diff line number
Diff line change
@@ -1,7 +1,7 @@
1
1
Isca Beginner's Guide
2
2
====================
3
3
4
-
Below is a list of reading and activites that will help you get comfortable using Isca. Be assured that your supervisor/tutor will not expect you to be fluent with this when you start using Isca, but it will help to have an idea of what to expect when you start running the model.
4
+
Below is a list of reading and activities that will help you get comfortable using Isca. Be assured that your supervisor/tutor will not expect you to be fluent with this when you start using Isca, but it will help to have an idea of what to expect when you start running the model.
5
5
6
6
This document is essentially a suggestion of signposts. With this kind of work, self-study and initiative is very important. It is up to you to go and research the topics until you feel comfortable.
7
7
@@ -112,7 +112,7 @@ Now you have made plots – or indeed any file you want to transfer between the
112
112
113
113
*SFTP (SSH File Transfer Protocol)* is one, it will work on all operating systems and is the easiest for windows. One way of using SFTP is with an *SFTP client*, many are available. One of them is `Cyberduck <https://cyberduck.io>`_. It will require setting up but it is fairly straight forward. These clients tend to have a GUI so you can just drag and drop the files you want to transfer. It is also possible to view and transfer files using the native file browser if you're using Linux or macOS, using their built-in functions to connect via SFTP.
114
114
115
-
Other option is to use a command line function, for example ``scp``. This is a secure file copy protol, which uses SSH. The usage is simple, for example on the computer you want to transfer the file to, type:
115
+
Other option is to use a command line function, for example ``scp``. This is a secure file copy protocol, which uses SSH. The usage is simple, for example on the computer you want to transfer the file to, type:
Copy file name to clipboardExpand all lines: docs/source/remote_access.rst
+6-6Lines changed: 6 additions & 6 deletions
Display the source diff
Display the rich diff
Original file line number
Diff line number
Diff line change
@@ -4,16 +4,16 @@ Remote Access
4
4
Overview
5
5
--------
6
6
This is a guide for how to edit remote files on a local text editor via port-
7
-
fowarding. By the end you should know:
7
+
forwarding. By the end you should know:
8
8
9
9
* how to quickly login to your remote server using an SSH config file
10
10
* how to edit remote files in a local text editor through ``rmate``
11
11
* how to edit remote python files in a local ``jupyter`` environment
12
12
13
-
Simplifying logins and port-fowarding
13
+
Simplifying logins and port-forwarding
14
14
-------------------------------------
15
15
If you use a Unix-based operating system on your personal computer, you can make use
16
-
of an SSH config file to create shortcuts to your frequently used remote computers. We will take this one step further to simplify port-fowarding, a method which allows a user to redirect data from a specified remote host and port, through a secure tunnel, to a specified local port. Port-fowarding is helfpul because it will enable us to edit remote files locally.
16
+
of an SSH config file to create shortcuts to your frequently used remote computers. We will take this one step further to simplify port-forwarding, a method which allows a user to redirect data from a specified remote host and port, through a secure tunnel, to a specified local port. Port-forwarding is helpful because it will enable us to edit remote files locally.
17
17
18
18
As an example without any fancy tricks, let's set up an SSH tunnel that maps ``localhost`` port 3039 on my local machine to 3039 on my remote machine (the number is arbitrary as long as its between 1024 and 49150): ``$ ssh -l localhost:3039:host:3039 user@host``. You will then be required to enter in your password. This is cumbersome to repeat everytime we log in. Our goal will be to shorten the command to : ``$ ssh hostalias`` and without having to enter in your password. We give some instructions below:
19
19
@@ -51,12 +51,12 @@ You should not be able to log in simply by typing ``$ ssh hostalias``. Congratul
51
51
Edit remote files in a local text editor using ``rmate``
As an alternative to remote-based text editors such as ``vi`` and ``emacs``, we can
54
-
use port-fowarding to set up a `local-based text editor like Atom <https://atom.io>`_ which includes features such as syntax highlighting and code completion. For instructions to install ``rmate`` on your **local machine**, `click here <https://github.com/textmate/rmate>`_. Then to specifically use Atom to edit remote files, `click here <https://atom.io/packages/remote-atom>`_. You will need to add the following line to your ``~/.ssh/config`` file: ``RemoteForward 52698 localhost:52698``.
54
+
use port-forwarding to set up a `local-based text editor like Atom <https://atom.io>`_ which includes features such as syntax highlighting and code completion. For instructions to install ``rmate`` on your **local machine**, `click here <https://github.com/textmate/rmate>`_. Then to specifically use Atom to edit remote files, `click here <https://atom.io/packages/remote-atom>`_. You will need to add the following line to your ``~/.ssh/config`` file: ``RemoteForward 52698 localhost:52698``.
55
55
56
56
57
57
Edit remote python files in a ``jupyter`` environment
The ``jupyter`` environment is a great environent for data exploration and integrating
59
+
The ``jupyter`` environment is a great environment for data exploration and integrating
60
60
your figures inline with your code. To open your first Jupyter notebook, log in to your **remote machine** and type: ``$ jupyter lab --no-browser --port=3039``.
61
61
62
62
To make it even quicker, you can type the following on your **local machine**: ``$ssh remotehost "jupyter lab --no-browser --port=3039``. This should function because of all the work we put in during the port forwarding section. To shorten this command, add an alias to your ``~/.bashrc`` file on your **local machine**. I personally use the alias ``rjlab``.
@@ -81,7 +81,7 @@ Your final ``~/.ssh/config`` file should look like this (making sure to replace
81
81
82
82
* Do not use this with VPN, use ithome aka hashbang as proxy. If the connection is interrupted you can still reconnect, assuming the jupyter process is still running. But make sure not to leave zombie jupyter processes with open ports on remote hosts!
83
83
84
-
* Remember the port numbers chosen are arbitary. If you choose the same number as someone else on your network, their files may open up on your computer and vice versa!
84
+
* Remember the port numbers chosen are arbitrary. If you choose the same number as someone else on your network, their files may open up on your computer and vice versa!
Copy file name to clipboardExpand all lines: exp/test_cases/README.md
+3Lines changed: 3 additions & 0 deletions
Display the source diff
Display the rich diff
Original file line number
Diff line number
Diff line change
@@ -6,6 +6,9 @@
6
6
`bucket_hydrology`
7
7
* As described in Isca paper (Vallis et al., 2017) but without q-fluxes
8
8
9
+
`column_test_case`
10
+
* Single column configuration of Isca. Please cite McKim et al. (2024, submitted) (<https://doi.org/10.22541/essoar.170904795.55675140/v1>) if you use the SCM.
11
+
9
12
`frierson`
10
13
* Control case of the so-called `Frierson model` described in e.g. <https://doi.org/10.1175/JAS3753.1>
0 commit comments