search

davidbrochart / jupyter_server_kernels_proxy

A Jupyter Server Extension Proxying Kernels.
MIT License
3 stars 1 forks source link

Transparently loading/saving the notebook files on the host machine #3

Open jpc opened 1 year ago

jpc commented 1 year ago

As far as I could tell the notebook files are held on the "target" machine, the same one where the kernel code is actually run. Wouldn't it be useful to keep the notebooks on the "host" machine (where JupyterLab is running) so they can be backed up more easily?

The use case I imagine right now is having short lived targets in a spot cloud VM, in k8s or inside an Android/iOS mobile app.

davidbrochart commented 1 year ago

As far as I could tell the notebook files are held on the "target" machine, the same one where the kernel code is actually run.

It shouldn't be the case, because the contents service is running on the "host" machine.