Ron Johnson wrote:
So why is Perl-SP-INSERT so much slower than Perl-SQL-INSERT?
(I can imagine that the SP code path would be longer, but since IO
is the slowest part of the system, I'm surprised that it's *that*
much slower.)
I'm guessing that since PG allows overloaded SP names, the slowness is
coming from resolving which SP to run. But that is just a guess. In my
environment, I don't *have* overloaded SPs, only a single version of a
given name. But when I was doing the conversion from Oracle, it took me
a couple tries to get the SP signatures correct, so I'm pretty sure PG
is still going through the resolution logic, even if you only have a
single instance of a given name.
--
Guy Rouillier
---------------------------(end of broadcast)---------------------------
TIP 9: In versions below 8.0, the planner will ignore your desire to
choose an index scan if your joining column's datatypes do not
match