Zeugswetter Andreas SB SD wrote:
>
> > > > > If they do not affect performance, then why have them off?
> > > >
> > > > I think Jan said 2-3%. If we can get autovacuum from it, it would be a
> > > > win to keep it on all the time, perhaps.
> > >
> > > Assuming that the statistics get updated:
> > >
> > > How often should the sats table be queried?
> > > What sort of configurability would be needed?
> >
> > You could wake up every few minutes and see how the values have changed.
> > I don't remember if there is a way to clear that stats so you can see
> > just the changes in the past five minutes. Vacuum the table that had
> > activity.
>
> I cannot envision querying the stats every 4 seconds, especially if the stats
> thread already has most of the info in hand.
>
> I still think, that for best results the vacuums should happen continuously
> for single pages based on a hook in wal or the buffer manager. Do I remember
> correctly, that the active page (the one receiving the next row) already has
> a strategy for slot reuse ? Maybe this strategy should be the followed more
> aggressively ?
>
> Seems the worst case is a few row table that permanently get updated,
> it should be possible to harness this situation with above method.
Perhaps the statistics thread can trigger a semaphore when it sees that vacuum
needs to be run?
