Content-Type: text/plain; charset=utf-8
Content-Transfer-Encoding: quoted-printable
From: Doug Reynolds <mav@wastegate.net>
Mime-Version: 1.0 (1.0)
Subject: Re: checking for a NULL date in a partitioned table kills performance
Date: Fri, 23 Aug 2024 15:42:16 +0000
Message-ID: 
 <010001917fe71552-589dbc9b-2c78-4a0d-8405-040db3cc09d7-000000@email.amazonses.com>
References: <1F2BA008-3732-4289-876A-EF6889474E08@elevated-dev.com>
In-Reply-To: <1F2BA008-3732-4289-876A-EF6889474E08@elevated-dev.com>
To: Pgsql-admin <pgsql-admin@lists.postgresql.org>
Feedback-ID: 
 ::1.us-east-1.L+CISTGIImABHDx8bBKKEgULZ36xzLwmtjped/xJNPU=:AmazonSES
Archived-At: 
 <https://www.postgresql.org/message-id/010001917fe71552-589dbc9b-2c78-4a0d-8405-040db3cc09d7-000000%40email.amazonses.com>
Precedence: bulk

The only difference is that you would be reading from one index instead of t=
wo, which could be more efficient.=20


Sent from my iPhone

> On Aug 23, 2024, at 11:19=E2=80=AFAM, Scott Ribe <scott_ribe@elevated-dev.=
com> wrote:
>=20
> =EF=BB=BF
>>=20
>> On Aug 23, 2024, at 5:39=E2=80=AFAM, Doug Reynolds <mav@wastegate.net> wr=
ote:
>>=20
>> You could try something really ugly where you make a function index that C=
OALESCEs the nulls to 1-1-1900 and use the COALESCE in the query.
>=20
> I don't see how that could be better than just creating a partial index on=
 it WHERE contract_date IS NULL--and anyway I'm sure you're right that no in=
dex would help. No matter what, it seems that sequential scans of all partit=
ions will be required since most rows have it null, and it's not even relate=
d to the partition key.
>=20