On Fri, Aug 15, 2025 at 12:24 PM Peter Geoghegan <pg@bowt.ie> wrote:
> Good news here: with Andres' bufmgr patch applied, the similar forwards scan
> query does indeed get more than 2x faster. And I don't mean that it gets
> faster on the randomized table -- it actually gets 2x faster with your
> original (almost but not quite entirely sequential) table, and your original
> query. This is especially good news because that query seems particularly
> likely to be representative of real world user queries.
BTW, I also think that Andres' patch makes performance a lot more
stable. I'm pretty sure that I've noticed that the exact query that I
just showed updated results for has at various times run faster
(without Andres' patch), due to who-knows-what.
FWIW, this development probably completely changes the results of many
(all?) of your benchmark queries. My guess is that with Andres' patch,
things will be better across the board. But in any case the numbers
that you posted before now must now be considered
obsolete/nonrepresentative. Since this is such a huge change.
--
Peter Geoghegan
