Re: Improving executor performance - tidbitmap

On 2016-07-14 20:41:21 -0700, Peter Geoghegan wrote:
> On Wed, Jul 13, 2016 at 8:06 PM, Andres Freund <andres(at)anarazel(dot)de> wrote:
> > I've quickly hacked up an alternative linear addressing hashtable
> > implementation. And the improvements are quite remarkable.
> >
> > Example Query:
> > EXPLAIN ANALYZE SELECT SUM(l_extendedprice) FROM lineitem WHERE (l_shipdate >= '1995-01-01'::date) AND (l_shipdate <= '1996-12-31'::date);
>
> > timing without analyze: 4136.425 4101.873 4177.441
> >
> > after:
>
> > timing without analyze: 2647.364 2674.456 2680.197
> >
> > as you can see the the time for the bitmap index scan goes from 2461.622
> > to 952.161.
>
> That's pretty great. I wonder what this would look like with a BRIN
> index, since l_shipdate looks like a good candidate for BRIN indexing.

Brin indexes IIRC always end up using tidbitmap.c, so the benefits
should be there as well ;)