From: | Ayush Vatsa <ayushvatsa1810(at)gmail(dot)com> |
---|---|
To: | PostgreSQL Hackers <pgsql-hackers(at)postgresql(dot)org> |
Subject: | Query regarding pg_prewarm extension |
Date: | 2024-12-13 10:46:16 |
Message-ID: | CACX+KaM05bqcJmg1wjGKOCHC5G7drjTd+PdRi=RjCQLqHhnxWA@mail.gmail.com |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-hackers |
Hi PostgreSQL Community,
I have a question regarding the use of the pg_prewarm() function [1]
in the pg_prewarm extension. The function requires a relation name
and a range of pages (e.g., pages 10 to 50) to be warmed by shifting
them from disk to the buffer cache.
How can I decide which range of pages to prewarm?
I assume that it is related to hot pages in the relation,
but how can I identify which pages are likely to be hot
before they are even in the buffer cache?
Additionally, since tuples within a page can move to
different pages over time (due to operations like VACUUM FULL or REINDEX),
how should I handle this when selecting the pages to prewarm?
Any insights would be greatly appreciated.
Regards,
Ayush Vatsa
[1] https://www.postgresql.org/docs/current/pgprewarm.html#PGPREWARM-FUNCS
From | Date | Subject | |
---|---|---|---|
Next Message | Alvaro Herrera | 2024-12-13 10:53:17 | Re: [PoC] Reducing planning time when tables have many partitions |
Previous Message | Alvaro Herrera | 2024-12-13 10:21:18 | Re: Difference in dump from original and restored database due to NOT NULL constraints on children |