Virtual Layer from SQL query is extremely slow

Question

I have two layers, one is a grid of circles and the other is all Census Block Groups (CBG) in the state of Florida. I created a Virtual Layer out of this code

SELECT CirclesFlorida.name, group_concat(cbgFl.FIPS)
FROM CirclesFlorida
LEFT JOIN cbgFl ON ST_Intersects(circlesFlorida.geometry, cbgFl.geometry)
GROUP BY circlesflorida.name

The purpose is for each circle to have a data column which lists every CBG that falls in that circle.

Is there a faster way to do this? Maybe DB Manager? Currently the query takes 20 minutes to run and another 5 minutes to just open the attribute table of the Virtual Layer.

• Inactivated Account: the db manager sql interface will use a sqlite database on the back-end to do the processing... I would try it out. I don't know if the sql layer is doing the same.
• Taras: As an alternative try to construct a model in the Graphical Modeler, it can also work with SQL queries, see this gis.stackexchange.com/a/360796/99589.
• Taras: Otherwise you may try running the Virtual Layer via Python console, check this gis.stackexchange.com/questions/227203/…
• Ben Hendel: @DPSSpatial the sqlite doesn't recognize the ".geometry" attribute as a column. Is it written in a different kind of SQL? How would I write it?
• Kay: Try to use _searchframe_ in your query! This helped me: lists.osgeo.org/pipermail/qgis-developer/2021-May/063582.html