Checkpoints
Warning
The Serverlesss mode will be discontinued on March 1, 2024.
A checkpoint in DataSphere is a saved state of a notebook. It includes cell code, output, and variable values, as well as project storage data.
Note
Checkpoints are not available in Dedicated mode as it does not allow saving a state.
You can create a maximum of 10 regular checkpoints per project. If a project already has 10 checkpoints, older checkpoints are automatically erased when creating a new checkpoint. To protect a checkpoint against deletion, pin it. You can pin a maximum of 10 checkpoints.
Note
You can fully use a checkpoint as a resource, that is, share it with other users and create other resources based on it, only after you pin it. You can learn more about working with checkpoints here.
Information about a checkpoint as a resource
A list of project checkpoints provides the following information about each checkpoint:
- The type of the operation after which the checkpoint was created, such as
Cell run
. - The name of the notebook where the cell is located, such as
Untitled.ipynb
, for a checkpoint created when running the code cell. - Tag of the checkpoint's additional content. For example, the
GIT
tag means that the cell contains a commit to the Git repository. - Checkpoint creation date in UTC
format, such asJuly 18, 2022, 14:23
.
To view detailed information about a checkpoint, click it in the list of project checkpoints.
Checkpoint details also include:
- Checkpoint name for a pinned checkpoint or the name of the operation that resulted in a checkpoint for non-pinned checkpoints.
- Unique resource ID.
- Name of the user who pinned the checkpoint.
For checkpoints created after running cells, the project cell code is available in the Cell code section. If a saved state contains variables, you can view their current values in the Variables tab.