Why are groupby and select queries excluded from caching by default?

is there anything special about these queries?

Hi Prashant,

I have just loaded the wikiticker data from druid quickstart.

I am even able to get the topN query result from backend which confirms that the druid data is loaded successfully.

But when I query using superset i get “No Data Returned”

Kindly find the attached screenshot.

Thanks and regards,

Dipesh Vora

The main reason is their result sets can be very large compared to other query types. So a small number of big groupBy/scan queries can quickly crowd out other query types from the cache. This PR is out to add a cap on cache entry size and enable caching for groupBy by default: https://github.com/apache/incubator-druid/pull/5108