Skip to content

Using the Python interpreter

Learning objectives

For teachers

Teaching goals are:

  • Learners have used the documentation of their HPC cluster
  • Learners have used the Python book
  • Learners have used the Python interpreter
  • Learners have loaded Python

Other goals are:

  • Verify that learners indeed have learned how to login
  • Verify that learners indeed have learned how to use the module system

Prior and feedback question:

  • What is Python?
  • What are features of the Python language?
  • What is an interpreted language?
  • What is a scripting language?
  • What is an interpreter?
  • What is an HPC cluster?
  • What are features of your HPC cluster?
  • What is a login node?
  • What is the rule on what to run on a login node?
  • How to run heavy Python calculations on your HPC cluster?
  • How to make your Python scripts reproducible?

Lesson plan:

  • 5 mins: prior knowledge
  • 5 mins: presentation
  • 15 mins: challenge
  • 5 mins: feedback

Why?

You want to make a computer run Python code, for some good reason. Here we do so!

To develop code in Python on your HPC clusters, one needs to:

  • Login to that HPC cluster, to be able to work on it
  • Load the Python module, to be able to use a modern Python interpreter
  • Use the Python interpreter, to run Python code

Luckily, the documentation of your HPC center should allow you to do this by yourself!

These exercises will ensure you are able to run a minimal Python program.

    flowchart TD
        python[[Python]]

        %% Give a white background to all nodes, instead of a transparent one
        classDef node fill:#fff,color:#000,stroke:#000

        subgraph sub_programming_language[Python as a language]
        interpreted_language[Interpreted language]
        programming_language[Programming language]
        scripting_language[Scripting language]
        interpreter[Interpreter]
        %% scripts[Scripts]
        %% text_files[Text files]
        end
        style sub_programming_language fill:#ccf,color:#000,stroke:#fcc


        subgraph sub_naiss[NAISS]
        your_hpc_center[Your HPC center]
        your_hpc_cluster[Your HPC cluster]
        modules[software modules]
        %% text_editors[Text editors]
        %%x_forwarding[X-forwarding]
        %%remote_desktop[Remote desktop]
        end
        style sub_naiss fill:#fcc,color:#000,stroke:#fcc

        python --> |is a| programming_language
        python --> |is a| interpreted_language
        python --> |is a| scripting_language

        %% programming_language --> |uses| text_files
        interpreted_language --> |has a| interpreter
        scripting_language --> |is a|interpreted_language
        %% scripting_language --> |runs| scripts
        %% scripts --> |are| text_files
        %% scripts --> |can use| command_line_arguments

        your_hpc_center --> |has| your_hpc_cluster
        your_hpc_cluster --> |uses| modules
        %%uppmax_clusters --> |have| text_editors
        %%uppmax_clusters --> |allow| x_forwarding
        %%uppmax_clusters --> |has| remote_desktop
        %% uppmax_modules --> |allow the use of| python_packages
        modules --> |allow the use of| interpreter

        %% text_editors --> |work on|text_files
        %%x_forwarding --> |allows|graphics
        %%remote_desktop --> |allows|graphics


        %% variables ~~~ uppmax

Exercises

Enjoy a video?

You can find a video with solutions to these exercises here

See the exercise procedure here.

Exercise 1: login to your HPC cluster

Go to the documentation of your HPC cluster and search for the login procedure(s).

Then answer these questions:

  • Login to your HPC cluster in any way
Answer

The documentation for the Rackham HPC cluster can be found at https://docs.uppmax.uu.se

Searching for, for example, 'Rackham login', you will find the UPPMAX page 'Login to Rackham':

ssh -X sven@rackham.uppmax.uu.se

Exercise 2: load the Python module

Go to the documentation of your HPC cluster and load the software module for Python.

Answer

The UPPMAX page on Python gives this solution:

module load python

Exercise 3: start the Python interpreter

Go to the documentation of your HPC cluster and start the Python interpreter.

Answer

On Rackham, do so by typing on a terminal:

python
Cannot get this to work?

If there is no time to get this fixed during the course, you are also use Python on your local computer.

Exercise 4: run a 'Hello world' program

Copy-paste the following code to the Python interpreter:

print('Hello, world!')

Press enter.

How does that look like?

Answer 
$ python
Python 3.12.3 (main, Nov  6 2024, 18:32:19) [GCC 13.2.0] on linux
Type "help", "copyright", "credits" or "license" for more information.
>>> print('Hello, world!')
Hello, world!
>>>

Congratulations, you've just run a 'Hello world' program 👍