I deal with some big (TB-level) partitioned tables where the application does a hybrid of querying both the master and the child tables directly. This clarification will be very useful.
Thank you again!
Nunks
----------
“Life beats down and crushes the soul and art reminds you that you have one.”
- Stella Adler
On Fri, Feb 19, 2016 at 11:32 AM, Tom Lane <tgl@sss.pgh.pa.us> wrote:
nunks <nunks.lol@gmail.com> writes: > What eludes me is that part of the documentation says we should analyze > each table individually, and another part says we can count on ANALYZE to > work on the entire inheritance tree...
Whole-tree stats are a different thing from per-table stats. What ANALYZE on a parent table produces is (1) stats for that table alone and (2) summary stats for the whole inheritance tree rooted at that table. If you want (3) stats for a child table alone then you need to ANALYZE that child table by name.