Re: Benchmarking partitioning triggers and rules

On 04/07/2015 07:49 PM, Tim Uckun wrote:
> I understand that there is overhead involved in parsing the strings and
> such. The amount of overhead was surprising to me but that's another
> matter. What I am really puzzled about is the difference between the
> statements
>
> EXECUTE 'INSERT INTO ' || quote_ident(partition_name) || ' SELECT
> ($1).*' USING NEW ;
>
> and
>
> EXECUTE 'INSERT INTO ' || quote_ident(partition_name) || ' VALUES
> (($1).*)' USING NEW ;

Offhand I would say because in the first case you are doing a SELECT and
in the second you are just doing a substitution.

>
> They both do string interpolation but one is significantly faster than
> the other. Is there a third and even faster way?
>
> I am using RDS so I can't really do stored procs in C.
>

--
Adrian Klaver
adrian(dot)klaver(at)aklaver(dot)com