Re: On columnar storage

Tom Lane wrote:
> Alvaro Herrera <alvherre(at)2ndquadrant(dot)com> writes:
> > Tom Lane wrote:
> >> Actually ... if you intend to allow column storage to work with inherited
> >> tables (and if you don't, you'd better have a darn good reason why not),
> >> I think you probably want to do this join insertion *after* inheritance
> >> expansion, so you can join child column stores only to the appropriate
> >> child heap table, and not to the entire inheritance tree.
>
> > Won't this cause issues to MergeAppend optimizations?
>
> Like what?

Well, as I understand, MergeAppend needs to know the sort order of the
child node, right? But that's available only on the relation RTE, not
on the colstore-join RTE. Though now that I think about it, maybe I can
push that info from the relation RTE to the colstore-join RTE, since I
know the ordering will be the same.

> And if there are such issues, why do you think you wouldn't be
> expected to solve them?

Precisely. If I simply reject having column stores in partitioned
tables, then I don't *need* to solve them. Later in the project, when
some planner hacker decides to join, I can ask them for advice on how to
tackle them.

--
Álvaro Herrera http://www.2ndQuadrant.com/
PostgreSQL Development, 24x7 Support, Remote DBA, Training & Services