Re-Using Datasets inside the Same SQL Dashboard

[mtreigelman]

Hi folks, I am creating a SQL dashboard and want to know if I can re-use datasets within the same dashboard. 

The screenshot below captures what I would like to do pretty well, but to summarize... I need to run an computationally expensive query and want to visualize those results in a figure on the dashboard. I then need to aggregate that table further to create additional figures. Is there a way I can call the first query again so that my SQL warehouse does not need to run that expensive first query twice? 

[Walter_C]

You should be able to do it by using static widgets as mentioned in docs https://docs.databricks.com/en/dashboards/index.html#static-widget-parameters  

[seans]

This still needs a solution. @Walter_C gave a reference to having two visualizations use the same dataset. This question (and my current problem) is about referencing a SQL dataset as if it was a view inside another SQL dataset in the dashboard. Dataset "my_agg" references dataset "my_raw" in the same dashboard (this is before any visualizations are defined, so parameters and widgets have no bearing yet). 

[sonynbcu]

Bumping this. I also have a use case where it would be beneficial to reference one dataset from another dataset.

[rskuntoji24]

Would appreciate a solution on this at the earliest as well. Thanks!

[mtreigelman]

@rskuntoji24 My current workaround has been to create the first query as a materialized view in my catalog, and then run a workflow daily to update the view. 

The downsides to this strategy is that it might not be practical for streaming data (not an issue for me personally), and creating a materialized view is saving the data that the query outputs somewhere (so storage requirements are increasing). I am not sure exactly where materialized view data is being stored, but I would guess somewhere on your cloud provider... perhaps someone like @Walter_C  could confirm.