MIME-Version: 1.0
From: =?UTF-8?B?RXJ0YW4gS8O8w6fDvGtvZ2x1?= <ertan.kucukoglu@gmail.com>
Date: Wed, 20 Aug 2025 07:47:43 +0300
Message-ID: <CAH2i4yf_M5NfX_kDu6_2Z+sgqLOZK9vCsF3EfmwZQyYFuAMxpg@mail.gmail.com>
Subject: Domains vs data types
To: pgsql-general@lists.postgresql.org
Content-Type: multipart/alternative; boundary="0000000000000158db063cc4adde"
Archived-At: <https://www.postgresql.org/message-id/CAH2i4yf_M5NfX_kDu6_2Z%2BsgqLOZK9vCsF3EfmwZQyYFuAMxpg%40mail.gmail.com>
Precedence: bulk

--0000000000000158db063cc4adde
Content-Type: text/plain; charset="UTF-8"

Hello,

I am using PostgreSQL 17.6.
I would like to learn if there is any benefit of using domains over data
types for table column definitions in terms of performance gain/loss.

For example I might have table defined as below
create table test (
  a integer,
  b integer,
  c integer,
  d varchar(5)
);

I might also have ame table defined as below

create domain aint integer;
create domain s5 varchar(5);

create table test_domain (
  a aint,
  b aint,
  c aint,
  d s5
);

Does the second table have any technical advantage/disadvantage over plain
data type definition?
Less metadata in memory? High metadata in memory? Less/increased disk space?

Thanks & Regards,
Ertan

--0000000000000158db063cc4adde
Content-Type: text/html; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable

<div dir=3D"ltr">Hello,<div><br></div><div>I am using PostgreSQL 17.6.</div=
><div>I would like to learn if there is any benefit of using domains over d=
ata types for table column definitions in=C2=A0terms of performance gain/lo=
ss.</div><div><br></div><div>For example I might have table defined as belo=
w</div><div><font face=3D"monospace">create table test (</font></div><div><=
font face=3D"monospace">=C2=A0 a integer,</font></div><div><font face=3D"mo=
nospace">=C2=A0 b integer,</font></div><div><font face=3D"monospace">=C2=A0=
 c integer,</font></div><div><font face=3D"monospace">=C2=A0 d varchar(5)</=
font></div><div><font face=3D"monospace">);</font></div><div><br></div><div=
>I might also have ame table defined as below</div><div><br></div><div><fon=
t face=3D"monospace">create domain aint integer;</font></div><div><font fac=
e=3D"monospace">create domain s5 varchar(5);</font></div><div><font face=3D=
"monospace"><br></font></div><div><font face=3D"monospace">create table tes=
t_domain (</font></div><div><font face=3D"monospace">=C2=A0 a aint,</font><=
/div><div><font face=3D"monospace">=C2=A0 b aint,</font></div><div><font fa=
ce=3D"monospace">=C2=A0 c aint,</font></div><div><font face=3D"monospace">=
=C2=A0 d s5</font></div><div><font face=3D"monospace">);</font></div><div><=
br></div><div>Does the second table have any technical advantage/disadvanta=
ge over plain data type definition?</div><div>Less metadata in memory? High=
 metadata in memory? Less/increased disk space?</div><div><br></div><div>Th=
anks &amp; Regards,</div><div>Ertan</div></div>

--0000000000000158db063cc4adde--