Horton

Re: walsender performance regression due to logical decoding on standby changes

Jeff Davis <pgsql@j-davis.com>

View thread View on postgr.es
From: Jeff Davis <pgsql@j-davis.com>
To: Andres Freund <andres@anarazel.de>, "Drouvot, Bertrand" <bertranddrouvot.pg@gmail.com>
Cc: pgsql-hackers@postgresql.org, Amit Kapila <amit.kapila16@gmail.com>, Robert Haas <robertmhaas@gmail.com>, Masahiko Sawada <sawada.mshk@gmail.com>
Date: 2023-05-17T19:46:33Z
Lists: pgsql-hackers 
On Wed, 2023-05-17 at 12:34 -0700, Andres Freund wrote:
> I understood Jeff's proposal to just have an early exit if there are
> no
> walsenders connected at all.

My suggestion was we early exit unless there is at least one *waiting*
walsender of the appropriate type. In other words, increment the
counter on entry to WalSndWait() and decrement on exit. I didn't look
in detail yet, so perhaps it's not easy/cheap to do that safely.

Regards,
	Jeff Davis

Commits

  1. Optimize walsender wake up logic using condition variables

  2. For cascading replication, wake physical and logical walsenders separately