Get the latest tech news
How we made Jupyter notebooks load faster
When we received feedback our Notebooks UI was taking too long too load, our engineers dove into ways to improve the developer experience — bringing some load times from 30 seconds down to less than one.
When you first opened a notebook, a call was made to our backend to assign you a container running JupyterLab; this included both the Jupyter Server, as well as the frontend and our custom extensions. We had 85% of it working, but it started to have so many rewrite rules and hacky mechanisms to cache HTML files (lab entry points), that we decided it wasn’t worth it to keep investing in it. As previously mentioned, when loading the iframe and starting JupyterLab, a bunch of requests are made to get extension settings, information about users, workspaces, kernels, sessions, etc.
Or read this on Hacker News
