Re: explain plans with information about (modified) gucs
От | Pavel Stehule |
---|---|
Тема | Re: explain plans with information about (modified) gucs |
Дата | |
Msg-id | CAFj8pRCmVGVK6rJjuWbOJe9LRQZfe1JELSONE_azhCKR6BWgKg@mail.gmail.com обсуждение исходный текст |
Ответ на | Re: explain plans with information about (modified) gucs (Tomas Vondra <tomas.vondra@2ndquadrant.com>) |
Список | pgsql-hackers |
út 1. 1. 2019 v 20:11 odesílatel Tomas Vondra <tomas.vondra@2ndquadrant.com> napsal:
On 1/1/19 6:48 PM, Pavel Stehule wrote:
>
>
> út 1. 1. 2019 v 18:39 odesílatel Tomas Vondra
> <tomas.vondra@2ndquadrant.com <mailto:tomas.vondra@2ndquadrant.com>> napsal:
>
> Attached is v4, changing how GUCs are picked for inclusion on the query
> plans. Instead of picking the GUCs based on group and/or explicitly, a
> new GUC_EXPLAIN flag is used for that.
>
> I went through GUCs defined in guc.c and marked those in QUERY_TUNING*
> groups accordingly, with the exception of default_statistics_target
> because that seems somewhat useless without showing the value used to
> actually analyze the table (and/or columns).
>
> I've also included a couple of other GUCs, that I find to be relevant:
>
> - parallel_leader_participation
> - max_parallel_workers_per_gather
> - max_parallel_workers
> - search_path
> - effective_io_concurrency
> - work_mem
> - temp_buffers
> - plan_cache_mode
>
>
> when plan_cache_mode is auto, you know maybe too less executed query.
> Maybe you can read somewhere if plan was custom or generic.
>
This patch is about showing GUCs, not such additional internal info.
Also, you'll see the plan actually used.
I understand to your goal, and I understand so collecting some data inside can be hard or impossible.
But if you collect a data important for understanding to planner behave/decision, then some important information is inside plancache too. - and now it is not visible from user space.
It is just my note - so not only GUC are interesting - nothing more.
>
> I think this covers the interesting GUCs pretty well, although perhaps I
> missed something.
>
>
> seq_page_cost, random_page_cost, from_collapse_limit,
> join_collapse_limit, ... enable_***
>
All these GUCs are included, of course.
ok - thank you for info.
Regards
Pavel
--
Tomas Vondra http://www.2ndQuadrant.com
PostgreSQL Development, 24x7 Support, Remote DBA, Training & Services
В списке pgsql-hackers по дате отправления: