pgv18: simple table scan take more time than pgv14

From: James Pang <jamespang886(at)gmail(dot)com>
To: pgsql-hackers(at)lists(dot)postgresql(dot)org
Subject: pgv18: simple table scan take more time than pgv14
Date: 2025-06-20 01:46:06
Message-ID: CAHgTRfc3q_Uv5LLKVi_D7RZjEHeC4YQg0wNG-5sZ+20fqFRVKg@mail.gmail.com
Views: Whole Thread | Raw Message | Download mbox | Resend email
Thread:
Lists: pgsql-hackers

same OS RHEL8, install pgv14.11 and pgv18.beta1_3, both installation by RPM
from pgdg, and use similar postgresql.conf.

14.11
postgres=# create table tt(a int primary key, b text);
CREATE TABLE
postgres=# do
postgres-# $$
postgres$# begin
postgres$# for counter in 1..1000000 loop
postgres$# insert into tt values(counter,'jamestest');
postgres$# end loop;
postgres$# end;
postgres$# $$;
DO

pgv14.11

postgres=# \timing on
Timing is on.
postgres=# \timing on
Timing is on.
postgres=# select * from tt where b ~~ 'a%';
a | b
---+---
(0 rows)

Time: 61.018 ms
postgres=# select * from tt where b ~~ 'a%';
a | b
---+---
(0 rows)

Time: 55.082 ms
postgres=# select * from tt where b ~~ 'a%';
a | b
---+---
(0 rows)

Time: 54.716 ms

postgres=# explain select * from tt where b ~~ 'a%';
QUERY PLAN
-------------------------------------------------------
Seq Scan on tt (cost=0.00..17906.00 rows=1 width=14)
Filter: (b ~~ 'a%'::text)
(2 rows)

Time: 0.447 ms

18.beta1
postgres=# \timing on
Timing is on.
postgres=# select * from tt where b ~~ 'a%';
a | b
---+---
(0 rows)

Time: 78.532 ms
postgres=# select * from tt where b ~~ 'a%'
postgres-# ;
a | b
---+---
(0 rows)

Time: 83.516 ms
postgres=# select * from tt where b ~~ 'a%';
a | b
---+---
(0 rows)

Time: 77.117 ms

psql (18beta1)
Type "help" for help.

postgres=# explain select * from tt where b ~~ 'a%';
QUERY PLAN
-------------------------------------------------------
Seq Scan on tt (cost=0.00..17906.00 rows=1 width=14)
Filter: (b ~~ 'a%'::text)
(2 rows)

Regards,

James

Responses

Browse pgsql-hackers by date

  From Date Subject
Next Message Michael Paquier 2025-06-20 02:05:37 Re: problems with toast.* reloptions
Previous Message Michael Paquier 2025-06-20 01:34:19 Re: Fixes inconsistent behavior in vacuum when it processes multiple relations