From: Mladen Gogala <gogala.mladen@gmail.com>
To: pgsql-admin@lists.postgresql.org
Subject: Re: Identifying "cold" data
Date: Mon, 28 Feb 2022 19:25:39 -0500
Message-ID: <29d1112e-0bc5-02d2-5f2f-7ed256edcea9@gmail.com> (raw)
In-Reply-To: <CAODZiv62tWEYUw4MdR_JBm8ZKJUnh8Dvh1-9N2dL5cehpiCHeg@mail.gmail.com>
References: <CAHs7QM87xwc=h6+p-WpBEvbbJKF8wp2a7NgP-0Qv3W-bgbRhnw@mail.gmail.com>
<9f1103cf-6073-4552-fb0e-84862bb226c3@gmail.com>
<CAODZiv62tWEYUw4MdR_JBm8ZKJUnh8Dvh1-9N2dL5cehpiCHeg@mail.gmail.com>
On 2/28/22 18:54, Keith Fiske wrote:
>
> On Thu, Feb 24, 2022 at 9:48 PM Ron <ronljohnsonjr@gmail.com> wrote:
>
> On 2/24/22 12:37 PM, Joseph Hammerman wrote:
>> Hi postgresql-admins,
>>
>> Has anyone put any thought or effort into figuring out how to
>> measure the total volume of data in a database against *how much
>> of it is hot*? I'm looking for some automatable approaches.
>> Similarly, is there a way to measure rarely queried columns, or
>> unused functions & triggers?
>
> https://www.postgresql.org/docs/9.6/monitoring-stats.html
>
> pg_stat_user_tables tells you how many records have been inserted,
> updated and deleted since the instance was started. It does not,
> though, say *which* records were updated,
>
>
> --
> Angular momentum makes the world go 'round.
>
>
> For monitoring data itself, I'm not sure of anything else built in
> other than what Ron shared with the stats table. If this is something
> you really need to monitor, perhaps look into adding a "changed_at"
> column to the tables that really need it and set a trigger to
> automatically update the timestamp of that column whenever the row is
> updated. Not sure how to track column usage.
>
> For tracking whether functions are used, you can look at the
> pg_stat_user_functions catalog
> (https://www.postgresql.org/docs/14/monitoring-stats.html#MONITORING-PG-STAT-USER-FUNCTIONS-VIEW),
> but as the documentation there notes, you do have to enable
> the "track_functions" parameter. I think this tracks trigger function
> usage as well, but would be good to test to make sure.
>
> --
> Keith Fiske
> Senior Database Engineer
> Crunchy Data - http://crunchydata.com
One way would be the proper table design, with TIMESTAMP column which
records the time of the last modification. Another way would be a table
trigger which would do time accounting for the table. You can implement
a "clock" algorithm which would set a "modified" column to 0 and then
wait for any application to modify the column and set it to 1. The
second hand of the clock would then set the "old" column. That is
roughly analogous to the way Unix and Linux paging systems work.
--
Mladen Gogala
Database Consultant
Tel: (347) 321-1217
https://dbwhisperer.wordpress.com
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Reply to all the recipients using the --to and --cc options:
reply via email
To: pgsql-admin@postgresql.org
Cc: gogala.mladen@gmail.com, pgsql-admin@lists.postgresql.org
Subject: Re: Identifying "cold" data
In-Reply-To: <29d1112e-0bc5-02d2-5f2f-7ed256edcea9@gmail.com>
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
This inbox is served by DDX for PostgreSQL; see mirroring instructions
for how to clone and mirror all data and code used for this inbox