Re: Seeking reason behind performance gain in 12 with HashAggregate

Поиск

Список

Период

Сортировка

От	Shira Bezalel
Тема	Re: Seeking reason behind performance gain in 12 with HashAggregate
Дата	13 января 2020 г. 20:34:04
Msg-id	CAE0KEwF-TZvn-Uar68Vem_jrKxoS=bzxmaBmxmFKFQ555-rg5Q@mail.gmail.com обсуждение исходный текст
Ответ на	Re: Seeking reason behind performance gain in 12 with HashAggregate (Justin Pryzby <pryzby@telsasoft.com>)
Ответы	Re: Seeking reason behind performance gain in 12 with HashAggregate (Michael Lewis <mlewis@entrata.com>)
Список	pgsql-performance

Дерево обсуждения

Hi Justin,

I'm seeing no difference in the query plan with JIT disabled in 12.1.

Thanks,

Shira

On Mon, Jan 13, 2020 at 8:42 AM Justin Pryzby <pryzby@telsasoft.com> wrote:

On Mon, Jan 13, 2020 at 08:29:05AM -0800, Shira Bezalel wrote:
> Here is the EXPLAIN (ANALYZE ON, BUFFERS ON) output from both systems:
>
> 9.6 plan <https://explain.depesz.com/s/W8HN>
> 12.1 plan <https://explain.depesz.com/s/lIRS>

> Is there something that was changed/improved in either 10, 11 or 12 that this
> behavior can be attributed to?

v12 has JIT enabled by default.
You can test if that's significant with SET jit=off.