Difference between revisions of "Getting started on Mac"

(Created page with "Mac operating system and Linux are both based on Unix. This means it is possible to connect to the supercomputer Gadi from the Mac straight out of the box. However, some funct...")
 
(NCI Virtual Desktop (VDI))
 
(11 intermediate revisions by the same user not shown)
Line 1: Line 1:
 +
[[Category:Setting up]]
 +
 
Mac operating system and Linux are both based on Unix. This means it is possible to connect to the supercomputer Gadi from the Mac straight out of the box. However, some functionalities are missing. We also recommend installing several software to improve on your experience.
 
Mac operating system and Linux are both based on Unix. This means it is possible to connect to the supercomputer Gadi from the Mac straight out of the box. However, some functionalities are missing. We also recommend installing several software to improve on your experience.
  
Line 5: Line 7:
 
For installation, you can go to the [https://www.xquartz.org/ XQuartz website] and follow the prompts or use [https://youtu.be/uS4zTqfwSSQ this video]
 
For installation, you can go to the [https://www.xquartz.org/ XQuartz website] and follow the prompts or use [https://youtu.be/uS4zTqfwSSQ this video]
  
== NCI Virtual Desktop (VDI) ==
+
== SSH configuration ==
NCI's virtual desktops (VDI) are an easy way to connect to NCI facilities from desktop computers. VDI provides a Linux desktop in a window, running on NCI's cloud
+
You will connect to remote servers (Gadi, VDI, Accessdev) using SSH. It is worth setting up SSH keys and a SSH configuration file. You can find detailed instructions on the [[CLEx_induction|induction page]] or in [https://youtu.be/AcP2FEFywXA our SSH training video].
  
Check NCI's VDI documentation for the current installation and setup instructions https://vdi.nci.org.au/help
+
== NCI Open OnDemand Service(OOD) ==
 +
NCI's Open OnDemand Service (OOD) is an easy way to connect to NCI facilities from desktop computers. OOD provides a Linux desktop in a browser and access to JupyterLab, running on NCI's cloud
  
Advantages:
+
Check NCI's OOD documentation for the current installation and setup instructions [https://opus.nci.org.au/display/OOD/Open+OnDemand+%28OOD%29+Service
  
    Access to NCI's /g/data filesystem
+
'''Advantages:'''
    Conda and Matlab (see NCI documentation) available
+
* Access to NCI's /g/data filesystem
    No compute time/SU charge
+
* Conda and Matlab (see NCI documentation) available
    Doesn't require a powerful home computer
+
* Doesn't require a powerful home computer
    When viewing graphical windows, connection can be faster than via X11 on slow networks
+
* When viewing graphical windows, connection can be faster than via X11 on slow networks
  
Disadvantages:
+
'''Disadvantages:'''
 +
* Can not access /scratch
  
    VDI servers are shared between multiple users
+
== Anaconda Python / Rstudio ==
  
== Anaconda Python / Rstudio ==
+
We recommend using [https://anaconda.org Anaconda] if you're working with Python on your own computer. This makes it easy to install packages like numpy and netCDF4, which depend on binary libraries. Anaconda uses the same system we use for the Conda environments at NCI. Anaconda also supports Rstudio if you work with R.
 +
 
 +
You can install the Mac native version. Unfortunately, if you keep the Mac default shell (zsh), after the installation is finished, you will need an additional step on Mac OS Catalina and newer. In a terminal, you will need to run the following command:
 +
* if you have installed with the graphical installer:
 +
 
 +
  bash -c "/opt/anaconda3/bin/conda init conda"
 +
* if you have installed with the command line installer:
 +
 
 +
  bash -c "~/anaconda3/bin/conda init conda"
 +
 
 +
If you have changed your shell to bash, the installation will work without the additional step.
 +
 
 +
You can check [https://www.youtube.com/watch?v=cfZEN3Ccd1E this video] for the instructions with the graphical installer. We recommend enabling the 'conda-forge' package repository, you will find a wider variety of packages there and they are generally well supported.
 +
 
 +
== VSCode ==
  
We recommend using Anaconda if you're working with Python on your own computer. This makes it easy to install packages like numpy and netCDF4, which depend on binary libraries. Anaconda uses the same system we use for the Conda environments at NCI. Anaconda also supports Rstudio if you work with R.
+
[https://code.visualstudio.com/ VS Code] is a nice code editor that can run on Windows, Mac and Linux. You can edit code locally, or use plugins to remotely connect to servers over SSH. It also integrates with Anaconda, letting you run Python programs in different environments.
  
You can install the Mac native version. We recommend enabling the 'conda-forge' package repository, you will find a wider variety of packages there and they are generally well supported.
+
== File transfers ==
 +
It is sometimes useful to transfer files between NCI servers and your local computer. You can do so via the command line with programs like rsync which is readily available. If you prefer a graphical interface, NCI advises you can install a program like [https://cyberduck.io/?l=en Cyberduck]. You can then use a SFTP connection. To access files on Gadi, it is preferable to use the data mover nodes: gadi-dm.nci.org.au for the server's name.

Latest revision as of 16:47, 16 August 2021


Mac operating system and Linux are both based on Unix. This means it is possible to connect to the supercomputer Gadi from the Mac straight out of the box. However, some functionalities are missing. We also recommend installing several software to improve on your experience.

XQuartz: required

You will need to install XQuartz. This software allows you to show remote windows from Gadi on your own screen. For installation, you can go to the XQuartz website and follow the prompts or use this video

SSH configuration

You will connect to remote servers (Gadi, VDI, Accessdev) using SSH. It is worth setting up SSH keys and a SSH configuration file. You can find detailed instructions on the induction page or in our SSH training video.

NCI Open OnDemand Service(OOD)

NCI's Open OnDemand Service (OOD) is an easy way to connect to NCI facilities from desktop computers. OOD provides a Linux desktop in a browser and access to JupyterLab, running on NCI's cloud

Check NCI's OOD documentation for the current installation and setup instructions [https://opus.nci.org.au/display/OOD/Open+OnDemand+%28OOD%29+Service

Advantages:

  • Access to NCI's /g/data filesystem
  • Conda and Matlab (see NCI documentation) available
  • Doesn't require a powerful home computer
  • When viewing graphical windows, connection can be faster than via X11 on slow networks

Disadvantages:

  • Can not access /scratch

Anaconda Python / Rstudio

We recommend using Anaconda if you're working with Python on your own computer. This makes it easy to install packages like numpy and netCDF4, which depend on binary libraries. Anaconda uses the same system we use for the Conda environments at NCI. Anaconda also supports Rstudio if you work with R.

You can install the Mac native version. Unfortunately, if you keep the Mac default shell (zsh), after the installation is finished, you will need an additional step on Mac OS Catalina and newer. In a terminal, you will need to run the following command:

  • if you have installed with the graphical installer:
 bash -c "/opt/anaconda3/bin/conda init conda"
  • if you have installed with the command line installer:
 bash -c "~/anaconda3/bin/conda init conda"

If you have changed your shell to bash, the installation will work without the additional step.

You can check this video for the instructions with the graphical installer. We recommend enabling the 'conda-forge' package repository, you will find a wider variety of packages there and they are generally well supported.

VSCode

VS Code is a nice code editor that can run on Windows, Mac and Linux. You can edit code locally, or use plugins to remotely connect to servers over SSH. It also integrates with Anaconda, letting you run Python programs in different environments.

File transfers

It is sometimes useful to transfer files between NCI servers and your local computer. You can do so via the command line with programs like rsync which is readily available. If you prefer a graphical interface, NCI advises you can install a program like Cyberduck. You can then use a SFTP connection. To access files on Gadi, it is preferable to use the data mover nodes: gadi-dm.nci.org.au for the server's name.