Fix the miss consideration of tuple_fraction during add_paths_to_append_rel

From: Andy Fan <zhihui(dot)fan1213(at)gmail(dot)com>
To: pgsql-hackers <pgsql-hackers(at)postgresql(dot)org>
Subject: Fix the miss consideration of tuple_fraction during add_paths_to_append_rel
Date: 2023-04-10 08:35:09
Message-ID: CAKU4AWqEnzhUTxopVhENC3vs6NnYV32+e6GSBtp1rAv0ZNX=mQ@mail.gmail.com
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-hackers

When I am working on "Pushing limit into subqueries of a union" [1], I
found we already have a great infrastructure to support this. For a query
like

subquery-1 UNION ALL subquery-2 LIMIT 3;

We have considered the root->tuple_fraction when planning the subqueries
without an extra Limit node as an overhead. But the reason it doesn't work
in my real case is flatten_simple_union_all flat the union all subqueries
into append relation and we didn't handle the root->tuple_fraction during
add_paths_to_append_rel.

Given the below query for example:
explain analyze
(select * from tenk1 order by hundred)
union all
(select * from tenk2 order by hundred)
limit 3;

Without the patch: Execution Time: 7.856 ms
with the patch: Execution Time: 0.224 ms

Any suggestion is welcome.

[1] https://www.postgresql.org/message-id/11228.1118365833%40sss.pgh.pa.us

--
Best Regards
Andy Fan

Attachment Content-Type Size
v1-0001-Considering-root-tuple_fraction-during-create_app.patch application/octet-stream 3.7 KB

Responses

Browse pgsql-hackers by date

  From Date Subject
Next Message Hayato Kuroda (Fujitsu) 2023-04-10 09:16:09 RE: [PoC] pg_upgrade: allow to upgrade publisher node
Previous Message Kyotaro Horiguchi 2023-04-10 08:35:00 eclg -C ORACLE breaks data