Re: Aggregation push down, reorder of join and group by

Hi,

On 2016/10/05 14:19, Chenxi Li wrote:
> Hello,
> 
> I'm reading some papers about aggregation push down like in "
> https://ub-madoc.bib.uni-mannheim.de/37228/1/main.pdf" and "
> http://www.vldb.org/conf/1995/P345.PDF". I think it is very useful but very
> complex to implement. In some complex queries, it can be a lot faster. Is
> there any plan to do this in the future?

Thanks for the links.  On a quick look, I think you are referring to the
following section in the document at the first link (and the title of the
paper at the second link):

4.2 Applying Eager Aggregation

I think someone is working on something like that.  Check out the
following discussion (currently inactive though):

* Partial Aggregation / GROUP BY before JOIN *
https://www.postgresql.org/message-id/flat/CAKJS1f9kw95K2pnCKAoPmNw%3D%3D7fgjSjC-82cy1RB%2B-x-Jz0QHA%40mail.gmail.com


There is also related work which is under active development:

* Aggregate Push Down - Performing aggregation on foreign server *

https://www.postgresql.org/message-id/flat/CAM2%2B6%3DW%3Dr_vh2gpccPJ4tc%3D%2BhFm4UGQsQdZgt4GiBMYVEk%2B5vg%40mail.gmail.com

In this case, the aggregation step is pushed all the way to the remote
server where the data resides.  It stands to avoid a good deal of
unnecessary data traffic across the network.

Thanks,
Amit