Backend shutdown time?
We know that backend startup time is something we'd like to cut down as
much as possible, but has anyone ever looked at backend *shutdown* time?
I'm sitting here watching Michael Devogelaere's benchmark with "top",
and despite the fact that there's only one client process at a time,
there's frequently two or three backends running. Sometimes four.
The only way I can see to explain this is that it takes a fair amount
of time for the backend to exit after the client disconnects.
Sample "ps -ef" run showing four active backends and one client:
tgl 17647 1 1 21:18 ? 00:02:31 postmaster -i -F
tgl 24844 17647 0 23:29 ? 00:00:00 postgres: userconsult_test userd
tgl 24852 17647 0 23:29 ? 00:00:00 postgres: userconsult_test userd
tgl 24856 17647 0 23:29 ? 00:00:00 postgres: userconsult_test userd
tgl 24859 17113 0 23:29 pts/1 00:00:00 /home/tgl/qmail/qmail-getpw alia
tgl 24860 17647 0 23:29 ? 00:00:00 postgres: userconsult_test userd
While it clearly takes some amount of time to clean out our entries in
the PROC array, etc, this trace suggests that the cost is a lot higher
than one would've expected. Anyone have an idea what's going on here?
(BTW, the above processes are 7.1.3, but I doubt 7.2 is better.)
regards, tom lane
I said:
While it clearly takes some amount of time to clean out our entries in
the PROC array, etc, this trace suggests that the cost is a lot higher
than one would've expected. Anyone have an idea what's going on here?
Oh, never mind, I figured it out: the backends are all nice'd because
they're background processes. As long as the client side wants cycles
(which it will, until it issues a query to the next backend) it's
hard for the exiting backend(s) to get any cycles at all.
regards, tom lane