If hashing names instead of using OIDs is too expensive for some workload,
then that workload would need to be able to turn statement hashing off. So
it needs to be optional, just like queryId is optionally computed today.
For many cases the extra overhead of hashing object names is small compared
to optimization time plus execution time.
-----
Jim Finnerty, AWS, Amazon Aurora PostgreSQL
--
Sent from: https://www.postgresql-archive.org/PostgreSQL-hackers-f1928748.html