The very concept of SSH has fascinated me from the day I learnt about it and it continues to do so every time I use it.
A tool like SSH (and it’s sibling SCP) opens so many doors for you as an innovator and a developer that it just cannot be ignored.
Clearly, it is quite a mature tool at this point and is already tightly integrated in most frameworks such as
If you are still not convinced, here is a simple example of how powerful even a simple ssh tunnel can be.
I have 2 machines with me
with jupyter notebook server running on the localhost of the mighty_machine.
Now, if I wanted to train a machine learning model on the mighty_machine and I was working on the puny_machine (because this is probably a laptop and I want to chill on the couch with it. Relatable no???). I could simply do the following:
ssh -L 8888:localhost:8888 <user_on_mighty_machine>@<host_of_mighty_machine>
And that is it!
You have successfully tunneled your jupyter notebook’s server from the mighty_machine to the puny_machine.
All you have to do now is go to the default address in your browser (on the puny_machine) and BAM! jupyter notebook with all the computing capacity of the mighty_machine at your service.
How cool and most importantly simple was that!
Jupyter notebook was just an example, the same works for any server.
Tunneling gives us so much freedom and also lets us develop cool new applications. (including remote-teleop of robots)
Note, I am assuming you have started your jupyter notebook on the default port and not configured it to start on 0.0.0.0 instead of localhost. In which case it would be available to you already at the IP of the server, but, the biggest caveat being that anybody in your network could tap into whatever you are working on unless you have setup a password to access the jupyter server as well (which is, in my opinion a lot of work).
Ideas? comments? suggestions for improvement?
Feel free to reach me on my E-mail