Querying my document with a (large) nested map is insanely slow

[akimotode]

Dear all,

I have a collection that hosts documents with some fields and a big nested map of smaller items.
doc =
field1:value
field2:value
field3:value
field4:map {map_item:{field id, field b, field c, ...}

When querying this construct with a simple "where field1 == valueX, I get insanely slow query responses.

For 20 document with about 500 map_items query time is upwards of 10secs. would anyone know why this is?

Kind regards,
Kai

[akimotode]

Here is a sample performance from the logs.

There is 18 docs with a total of 4700 map_items, somewhat unevenly distributed. Some have 30, one has about 1000...

---

--------> Executing firestore query took 12103 milliseconds to execute
--------> Found 18 annotation documents
--------> Found 4707 annotations

I am on a 1GB fiber connection and the geo-location is on the same continent. I can pull down 500mb of python libraries faster than these 18 docs from a single collection.

Attached an image of what each map_item looks like.