I just had a thought- I know some of these tables are in need of a vacuuming.
Could it be that the dump is dumping a bunch of garbage that the restore has
to sift through on the restore? I don't know enough details to know if this
is a dumb thought or not.
The restore to RDS took roughly the same amount of time. My next move is to
try on a fast instance store, and also do a postgres 9 restore of a pure SQL
dump, but that won't really be a great test since I use custom format. I'm
assuming here that I can't take the custom dump from 9.2 and apply it to
9.0, or can I?
--
View this message in context:
http://postgresql.1045698.n5.nabble.com/Incredibly-slow-restore-times-after-9-0-9-2-upgrade-tp5824701p5825052.html
Sent from the PostgreSQL - performance mailing list archive at Nabble.com.
