Re: a wrong index choose when statistics is out of date - Mailing list pgsql-hackers

From Andrei Lepikhov
Subject Re: a wrong index choose when statistics is out of date
Date
Msg-id 085b3945-84fe-4f54-b29b-cc878a561392@postgrespro.ru
Whole thread Raw
In response to Re: a wrong index choose when statistics is out of date  (David Rowley <dgrowleyml@gmail.com>)
Responses Re: a wrong index choose when statistics is out of date
List pgsql-hackers
On 4/3/2024 12:33, David Rowley wrote:
> [1]
https://www.postgresql.org/message-id/CAApHDvo2sMPF9m%3Di%2BYPPUssfTV1GB%3DZ8nMVa%2B9Uq4RZJ8sULeQ%40mail.gmail.com
Thanks for the link!
Could we use the trick with the get_actual_variable_range() to find some 
reason and extrapolate histogram data out of the boundaries when an 
index shows us that we have min/max outside known statistics?
Because it would be used for the values out of the histogram, it should 
only add an overhead with a reason.

-- 
regards,
Andrei Lepikhov
Postgres Professional




pgsql-hackers by date:

Previous
From: "Andrey M. Borodin"
Date:
Subject: Re: Tidy fill hstv array (src/backend/access/heap/pruneheap.c)
Next
From: "Andrey M. Borodin"
Date:
Subject: Re: Add Index-level REINDEX with multiple jobs