- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
05-26-2024 08:40 PM
When we create a Lakeview dashboard, the visuals show truncated data. I want to create a dashboard using the entire dataset because the charts do not display the exact values needed for accurate analysis.
Accepted Solutions
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
05-27-2024 12:39 PM
Hello @Akshay_Petkar ,
The FrontEnd (FE) rendering is a limit that determines what data we render on the FE that has been returned by the BackEnd (BE). Render all button allows you to render all the data on the FE that has been returned by the BE.
Render all is available in interactive use cases like the Notebook or the Notebook Dashboard. Jobs runs are not interactive by definition, so there is no render all. A job runs, completes and has results. An individual run can’t be updated.
Every visualization on a Notebook that is run via a Job will either use the results from the table tab (FE aggregation w/ no cost to the customer) or it will run a separate command over all data (BE aggregation w/ a cost to the customer).
We decide between FE aggregation and BE aggregation based on how the Notebook asset was last run. If you would like this job to always BE aggregate the visualization so that it runs over all data, then you should make sure that Notebook that the job is based on does the same.
Raphael Balogo
Sr. Technical Solutions Engineer
Databricks
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
05-27-2024 12:39 PM
Hello @Akshay_Petkar ,
The FrontEnd (FE) rendering is a limit that determines what data we render on the FE that has been returned by the BackEnd (BE). Render all button allows you to render all the data on the FE that has been returned by the BE.
Render all is available in interactive use cases like the Notebook or the Notebook Dashboard. Jobs runs are not interactive by definition, so there is no render all. A job runs, completes and has results. An individual run can’t be updated.
Every visualization on a Notebook that is run via a Job will either use the results from the table tab (FE aggregation w/ no cost to the customer) or it will run a separate command over all data (BE aggregation w/ a cost to the customer).
We decide between FE aggregation and BE aggregation based on how the Notebook asset was last run. If you would like this job to always BE aggregate the visualization so that it runs over all data, then you should make sure that Notebook that the job is based on does the same.
Raphael Balogo
Sr. Technical Solutions Engineer
Databricks
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
05-27-2024 08:44 PM
Hello @Raphael Balogo
Thank you for the detailed explanation. I am referring to the SQL Lakeview Dashboard, where we select visualizations on the canvas and add fields to them. I would like to know if there is a way to display the entire dataset in a visualization on the Lakeview Dashboard, or if the data is always truncated.
Thank you,
Akshay Petkar.

