Re: Syncrep and improving latency due to WAL throttling

On Fri, Jan 27, 2023 at 2:03 PM Alvaro Herrera <alvherre(at)alvh(dot)no-ip(dot)org> wrote:
>
> On 2023-Jan-27, Bharath Rupireddy wrote:
>
> > Looking at the patch, the feature, in its current shape, focuses on
> > improving replication lag (by throttling WAL on the primary) only when
> > synchronous replication is enabled. Why is that? Why can't we design
> > it for replication in general (async, sync, and logical replication)?
> >
> > Keeping replication lag under check enables one to provide a better
> > RPO guarantee
>
> Hmm, but you can already do that by tuning walwriter, no?

IIUC, one can increase wal_writer_delay or wal_writer_flush_after to
control the amount of WAL walwriter writes and flushes so that the
walsenders will get slower a bit thus improving replication lag. If my
understanding is correct, what if other backends doing WAL writes and
flushes? How do we control that?

I think tuning walwriter alone may not help to keep replication lag
under check. Even if it does, it requires manual intervention.

--
Bharath Rupireddy
PostgreSQL Contributors Team
RDS Open Source Databases
Amazon Web Services: https://aws.amazon.com