On Mon, Feb 15, 2021 at 11:56 PM Andrey Borodin <x4mmm@yandex-team.ru> wrote:
> > 16 янв. 2021 г., в 03:07, Alvaro Herrera <alvherre@alvh.no-ip.org> написал(а):
> > Andrey Borodin already has a patch to change the behavior for
> > multixact, which is something we should perhaps also do. I now notice
> > that they're also reporting a bug in that thread ... sigh
>
> I've tried in that thread [0] to do more intelligent optimisation than just increase number of buffers.
> Though, in fact bigger memory had dramatically better effect that lock tricks.
>
> Maybe let's make all SLRUs buffer sizes configurable?
+1
I got interested in the SLRU sizing problem (the lock trick and CV
stuff sounds interesting too, but I didn't have time to review that in
this cycle). The main problem I'm aware of with it is the linear
search, so I tried to fix that. See Andrey's thread for details.