Re: GSOC - TOAST'ing in slices

George,

* George Papadrosou (gpapadrosou(at)gmail(dot)com) wrote:
> Stephen, you mentioned PostGIS, but the conversation seems to lean towards JSONB. What are your thoughts?

Both are important. I brought up PostGIS specifically because it's an
external project which could benefit from this work and to explain that
PostgreSQL can be extended to beyond the built-in data types. Focusing
on JSONB to start seems alright but keep in mind that we'll want to have
a way to let PostGIS do whatever it is we do for JSONB in core.

> Also, if I am to include some ideas/approaches in the proposal, it seems I should really focus on understanding how a specific data type is used, queried and indexed, which is a lot of exploring for a newcomer in postgres code.

This is true, but, really, the sooner the better. :) While it's not a
small amount to go through it's also really pretty clean code, in
general, so hopefully you won't have too much trouble. I would
recommend jumping on irc.freenode.net to the #postgresql channel where
you can find a number of PostgreSQL hacker who are generally quite happy
to answer specific questions you may have as you go through the code.

> In the meanwhile, I am trying to find how jsonb is indexed and queried. After I grasp the current situation I will be to think about new approaches.

I would suggest you make sure that you first understand how TOAST works
generally and review the on-disk storage format before jumping in to try
and understand jsonb indexing and queries. Would be good to also
understand how the PGLZ compression works.

Thanks!

Stephen