Re: Reducing power consumption on idle servers

Hi,

> Replication never got the memo, so power consumption on an
> idle server is not very effective on standby or logical subscribers.
> The code and timing for hibernation is also different for each worker,
> which is confusing.

Agree, this patch makes it easier to understand the hibernation
behavior of various workers.

> 1. Standardize the hibernation time at 60s, using a #define
> HIBERNATE_DELAY_SEC 60

I notice in patch 3 HIBERNATE_DELAY_SEC has been increased to 300
seconds, what’s the reasoning behind it? Is longer hibernation delay
better? If so can we set it to INT_MAX (the max timeout allowed by
WaitLatch()) in which case a worker in hibernation only relies on
wakeup? I think it would be nice to run experiments to verify that the
patch reduces power consumption while varying the value of
HIBERNATE_DELAY_SEC.

Regards,
Zheng Li
Amazon RDS/Aurora for PostgreSQL

On Thu, Mar 3, 2022 at 5:49 PM Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us> wrote:
>
> Jim Nasby <nasbyj(at)amazon(dot)com> writes:
> > I'm wondering if it'd be worth linking autovac wakeup from a truly idle
> > state to the stats collector. If there's no stats messages coming in
> > clearly there's nothing new for autovac.
>
> That seems pretty scary in the current system design, where the
> stats collector is intentionally not 100% reliable (and sometimes,
> less intentionally, it fails completely). If we get to a place
> where we're willing to bank on stats being collected 100% of the
> time, it might make sense.
>
> regards, tom lane
>
>