Re: Avoid detoast overhead when possible

Nikita Malakhov <hukutoc(at)gmail(dot)com> writes:

> Hi,
>
> Hmmm, I've checked this patch and can't see performance difference on a large
> (20000 key-value pairs) json, using toasted json column several times makes no
> difference between current implementation on master (like queries mentioned above).
>
> Maybe I'm doing something wrong?

Could you try something like below? (set jit to off to turn on this
feature). Or could you tell me the steps you used? I also attached the
setup.sql at the begining of this thread.

select big->'1', big->'2', big->'3', big->'5', big->'10' from b;
QUERY PLAN
---------------------------------------------------------------
Seq Scan on b (actual time=1.731..1577.911 rows=1001 loops=1)
Planning Time: 0.099 ms
Execution Time: 1578.411 ms
(3 rows)

set jit to off;

select big->'1', big->'2', big->'3', big->'5', big->'10' from b;
QUERY PLAN
--------------------------------------------------------------
Seq Scan on b (actual time=0.417..309.937 rows=1001 loops=1)
Planning Time: 0.097 ms
Execution Time: 310.255 m

(I used 'jit=off' to turn on this feature just because I'm still not
ready for JIT code.)

--
Best Regards
Andy Fan