On Sun, Aug 1, 2021 at 5:41 PM Thomas Munro <thomas.munro@gmail.com> wrote:
> On Fri, Jul 30, 2021 at 12:34 PM John Naylor
> <john.naylor@enterprisedb.com> wrote:
> > I got around to getting a benchmark together to serve as a starting point. I based it off something I got from the
archives,but don't remember where (I seem to remember Tomas Vondra wrote the original, but not sure). To start I just
usedtypes that were there already -- int, text, numeric. The latter two won't be helped by this patch, but I wanted to
keepsomething like that so we can see what kind of noise variation there is. I'll probably cut text out in the future
andjust keep numeric for that purpose.
>
> Thanks, that's very useful.
If somebody wants to get a sense of what the size hit is from all of
these specializations, I can recommend the diff feature of bloaty:
https://github.com/google/bloaty/blob/master/doc/using.md#size-diffs
Obviously you'd approach this by building postgres without the patch,
and diffing that baseline to postgres with the patch. And possibly
variations of the patch, with less or more sort specializations.
--
Peter Geoghegan