Shared resource model in code workspaces

When multiple users are using a singular Jupyter workspace with multiple notebooks/branches am I right in assuming all these users/resources share the fixed resources of the profile.

For example if the memory profile was 8gb would all active processes in various files, from all users, have to stay under 8gb?

Hey OliverC - thanks for the question and welcome to the Palantir Developer Community!

When working in a Jupyter workspace, resources aren’t shared across multiple users - each user gets their own session and own memory, which is allocated with the memory profile selected in their workspace settings. In fact, each users session is completely isolated from other users, so that you have your own view of your workspace that other’s can’t see.

When you set 8gb of memory in your own individualized workspace, it means your entire Jupyter instance (e.g. all processes running across notebook files, terminal processes, etc) has access to the pool of 8gb. You can think of it as having your own computer which is allocated 8gb of memory, where all your processes pull from the same resource pool.

Hope this helps answer your question

1 Like