Overview of notebooks

Learn about the building blocks of a notebook, how notebooks can help show how data is processed, and discover some common use cases.

To use different notebook cell types, controls, and uses to process data, see:

Notebook concepts

Using notebooks, you can:

  • Create documents with live code, equations, visualizations, and explanatory notes.
  • Create dashboard cells or scheduled tasks.
  • Clean and downsample data.
  • Build runbooks.
  • Document data processing steps.

Notebook controls

The following options appear at the top of each notebook.

Preview/Run mode

  • Select Preview (or press Control+Enter) to display results of each cell without writing data. Helps to verify that cells return expected results before writing data.
  • Select Run mode displays results of each cell and writes data to the selected bucket.

Presentation mode

Display notebooks in full screen with Presentation mode, hiding the left and top navigation menus so only the cells appear. This mode might be helpful, for example, for stationary screens dedicated to monitoring visualizations.

Local or UTC timezone

Click the timezone dropdown list to select a timezone to use for the dashboard. Select either the local time (default) or UTC.

Time range

Select from the options in the dropdown list or select Custom Time Range to enter a custom time range with precision up to nanoseconds.

Notebook cell types

The following cell types are available for your notebook:


At least one input cell is required in a notebook for other cells to run.

  • Metric Selector: Filter out your data using a single list.
  • Query Builder: Filter out your data using lists separated by group keys.


  • Flux Script: Modify your data with Flux Script. __PREVIOUS_RESULT__ refers to your previous inputs from the Metric Selector.
  • Downsample: Downsample data through aggregates.


  • Column Editor: Modify column visibility and naming.
  • Markdown: Create explanatory notes or other information for yourself or one of your team members.
  • Visualization: Represents your data in visual formats such as graphs or charts.


  • Notification: Set up alerts.
  • Output to Bucket: In Preview mode, this cell represents what would be written if the data was going to be committed to a bucket.
  • Output: Schedule: Run tasks at specific times.

Select your region

Upgrade to InfluxDB Cloud or InfluxDB 2.0!

InfluxDB Cloud and InfluxDB OSS 2.0 ready for production.