From: Rui DeSousa <rui@crazybean.net>
To: Thomas Kellerer <shammat@gmx.net>
Cc: Pgsql-admin <pgsql-admin@lists.postgresql.org>
Subject: Re: Transaction Size in PostgreSQL
Date: Tue, 7 Dec 2021 01:59:34 -0500
Message-ID: <5681E7B8-DCF4-48B6-B32F-6895670BB000@crazybean.net> (raw)
In-Reply-To: <17b19213-1588-dca9-b009-8913b8d1401f@gmx.net>
References: <CAGH8HirGbc_=EU3GYBWQYnofWU2y_ygjP=1gzp-j6wmE3CSMtQ@mail.gmail.com>
<CAKFQuwajBNigQZ=mU3_hG_-3uNcAzk6j_hPRSH=YDRU_ZKcZUA@mail.gmail.com>
<CAGH8HiqXBToqe_EVqCxPaV1ZP+vKA9hQDvLEFE-UYRkR+RzwEw@mail.gmail.com>
<17b19213-1588-dca9-b009-8913b8d1401f@gmx.net>
> On Dec 7, 2021, at 1:41 AM, Thomas Kellerer <shammat@gmx.net> wrote:
>
> "Massive data changes" in a single transaction aren't really a problem.
> In my experience a single transaction changing a lot of rows is more
> efficient than multiple transactions changing fewer rows.
That might work on a idle system but I would have to disagree with that statement. In theory it may be more efficient but it practice large long running transactions cause replication lag issues as well as vacuuming issues by holding on to older xmin value. Basically, they are bad for high transaction systems.
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: rui@crazybean.net, shammat@gmx.net, pgsql-admin@lists.postgresql.org
Subject: Re: Transaction Size in PostgreSQL
In-Reply-To: <5681E7B8-DCF4-48B6-B32F-6895670BB000@crazybean.net>
* 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