Re: [GENERAL] Problems with Greatest - Mailing list pgsql-general

From Thomas Kellerer
Subject Re: [GENERAL] Problems with Greatest
Date
Msg-id o83j8r$31t$1@blaine.gmane.org
Whole thread Raw
In response to Re: [GENERAL] Problems with Greatest  (Teddy Schmitz <teddy.schmitz@actuallywedo.com>)
List pgsql-general
Teddy Schmitz schrieb am 16.02.2017 um 05:38:
> As a quick follow up I just did an explain on the query,
>
>
> Aggregate  (cost=258007258.87..258007258.88 rows=1 width=8)
> ->  Nested Loop  (cost=0.00..184292254.83 rows=14743000807 width=16)
> ->  Seq Scan on t1  (cost=0.00..3796.41 rows=263141 width=8)
> ->  Materialize  (cost=0.00..1088.40 rows=56027 width=8)
> ->  Seq Scan on t2  (cost=0.00..808.27 rows=56027 width=8)
>
>
> It seems it has to do a loop on 14 billion rows?  Can someone explain why this would happen?

You are not joining those two tables.

"from t1, t2" is a cross join that will calculate the cartesian product of both tables

pgsql-general by date:

Previous
From: Teddy Schmitz
Date:
Subject: Re: [GENERAL] Problems with Greatest
Next
From: Alessandro Baggi
Date:
Subject: Re: [GENERAL] postgresql how to duplicate rows in result.