Re: Performance degradation with CTEs, switching from PG 11 to PG 15 - Mailing list pgsql-performance

From Andreas Kretschmer
Subject Re: Performance degradation with CTEs, switching from PG 11 to PG 15
Date
Msg-id 211a58a6-5741-a0a2-db97-0e79f75d9d4f@a-kretschmer.de
Whole thread Raw
In response to Performance degradation with CTEs, switching from PG 11 to PG 15  (Jean-Christophe Boggio <postgresql@thefreecat.org>)
Responses Re: Performance degradation with CTEs, switching from PG 11 to PG 15  (Jean-Christophe Boggio <postgresql@thefreecat.org>)
List pgsql-performance

Am 22.11.23 um 12:38 schrieb Jean-Christophe Boggio:
>
>
> Also, adding "materialized" to both "withcwrack" and "withcwrack0" 
> CTEs gets the result in acceptable timings (a few seconds). The 
> problem with this is that we have some clients with older versions of 
> PG and I guess blindly adding the "materialized" keyword will cause 
> errors.
>

yeah, prior to 11 CTEs are a optimizer barrier. You can try to rewrite 
the queries to not using CTEs - or upgrade. If i were you i would upgrade.


Regards, Andreas

-- 
Andreas Kretschmer - currently still (garden leave)
Technical Account Manager (TAM)
www.enterprisedb.com




pgsql-performance by date:

Previous
From: Jean-Christophe Boggio
Date:
Subject: Re: Performance degradation with CTEs, switching from PG 11 to PG 15
Next
From: Jean-Christophe Boggio
Date:
Subject: Re: Performance degradation with CTEs, switching from PG 11 to PG 15