Re: [BUG]: the walsender does not update its IO statistics until it exits

On Fri, Feb 28, 2025 at 02:41:34PM +0900, Michael Paquier wrote:
> With smaller records, the loop can become hotter, can't it? Also,
> there can be a high number of WAL senders on a single node, and I've
> heard of some customers with complex logical decoding deployments with
> dozens of logical WAL senders. Isn't there a risk of having this code
> path become a point of contention? It seems to me that we should
> benchmark this change more carefully, perhaps even reduce the
> frequency of the report calls.

One idea here would be to have on a single host one server with a set
of N pg_receivewal processes dumping their WAL segments into a tmpfs,
while a single session generates a bunch of records with a minimal
size using pg_logical_emit_message(). Monitoring the maximum
replication lag with pg_stat_replication and looking at some perf
profiles of the cluster should show how these stats reports affect the
replication setup efficiency.
--
Michael