I have a unique constraint on two columns of a supermassive table (est. 1.7
bn rows) that are the only way the table's ever queried - and it's
blindingly fast: 51ms to retrieve any single row even non-partitioned.
Anyway: Right now statistics on the two unique constrained columns are set
to 200 each (database-wide default is 100), and what I'm wondering is, since
the unique constraint already covers the whole table and all rows in
entirety, is it really necessary for statistics to be set that high on
those? Or does that only serve to slow down inserts to that table?
--
View this message in context:
http://postgresql.1045698.n5.nabble.com/statistics-target-for-columns-in-unique-constraint-tp5755256.html
Sent from the PostgreSQL - performance mailing list archive at Nabble.com.