From: Jim Nasby <jnasby@upgrade.com>
Message-Id: <0AA66026-3707-4252-B97A-86CC65F1F22E@upgrade.com>
Content-Type: multipart/alternative;
	boundary="Apple-Mail=_DFA44C2C-12DB-4491-89D6-2CEE4048254C"
Mime-Version: 1.0 (Mac OS X Mail 16.0 \(3826.400.131.1.6\))
Subject: Re: Built-in Raft replication
Date: Wed, 23 Apr 2025 11:48:46 -0500
In-Reply-To: <CAKAnmm+QJnL_APyKhM3hjkOv9DXjWaqJxB2nFDuNH9JqVdTfXA@mail.gmail.com>
Cc: Ashutosh Bapat <ashutosh.bapat.oss@gmail.com>,
 Andrey Borodin <x4mmm@yandex-team.ru>,
 Tom Lane <tgl@sss.pgh.pa.us>,
 Konstantin Osipov <kostja.osipov@gmail.com>,
 Nikolay Samokhvalov <nik@postgres.ai>,
 PostgreSQL Hackers <pgsql-hackers@lists.postgresql.org>
To: Greg Sabino Mullane <htamfids@gmail.com>
References: <Z_1Cq7JvabsFYjQo@ark>
 <CAKAnmmJxGHcQmzXqEQ-nppZjXyH7i5W_7Cz8JNMb2+6c5+D+og@mail.gmail.com>
 <CAM527d9+eGvdih1GxoPDJ2WYs6eitf39oNTxGxnCK1PgF=A8wg@mail.gmail.com>
 <1798838.1744759182@sss.pgh.pa.us>
 <20FB597F-641F-48F8-8428-D8DDBA802D58@yandex-team.ru>
 <CAExHW5udz081Unx5qcLZ89DnR4cp6OhW0TCoco3+AEkGw480uQ@mail.gmail.com>
 <212D5973-FDD0-4CF5-BCD0-2760EC319DF3@yandex-team.ru>
 <CAExHW5tq8ShEfboFa52wDMWmyaW+6kX_6x8MZ-0TP65tOhJ1wQ@mail.gmail.com>
 <CAKAnmm+QJnL_APyKhM3hjkOv9DXjWaqJxB2nFDuNH9JqVdTfXA@mail.gmail.com>
Archived-At: <https://www.postgresql.org/message-id/0AA66026-3707-4252-B97A-86CC65F1F22E%40upgrade.com>
Precedence: bulk


--Apple-Mail=_DFA44C2C-12DB-4491-89D6-2CEE4048254C
Content-Transfer-Encoding: quoted-printable
Content-Type: text/plain;
	charset=utf-8

On Apr 16, 2025, at 2:29=E2=80=AFPM, Greg Sabino Mullane =
<htamfids@gmail.com> wrote:
>=20
> On Wed, Apr 16, 2025 at 2:18=E2=80=AFAM Ashutosh Bapat =
<ashutosh.bapat.oss@gmail.com <mailto:ashutosh.bapat.oss@gmail.com>> =
wrote:
>> Users find it a waste of resources to deploy 3 big PostgreSQL =
instances just for HA where 2 suffice even if they deploy 3 lightweight =
DCS instances. Having only some of the nodes act as DCS and others =
purely PostgreSQL nodes will reduce waste of resources.
>=20
> A big problem is that putting your DCS into Postgres means your whole =
system is now super-sensitive to IO/WAL-streaming issues, and a busy =
database doing database stuff is going to start affecting the DCS stuff. =
 With three lightweight DCS servers, you don't really need to worry =
about how stressed the database servers are. In that way, I feel etcd et =
al. are adhering to the unix philosophy of "do one thing, and do it =
well.=E2=80=9D


=E2=80=A6 unless we added multiple WAL streams. That would allow for =
splitting WAL traffic across multiple devices as well as providing =
better support for configurations that don=E2=80=99t replicate the =
entire cluster. The current situation where delayed replication of a =
single table mandates retention of all the WAL for the entire cluster is =
less than ideal.


--Apple-Mail=_DFA44C2C-12DB-4491-89D6-2CEE4048254C
Content-Transfer-Encoding: quoted-printable
Content-Type: text/html;
	charset=utf-8

<html><head><meta http-equiv=3D"content-type" content=3D"text/html; =
charset=3Dutf-8"></head><body style=3D"overflow-wrap: break-word; =
-webkit-nbsp-mode: space; line-break: after-white-space;">On Apr 16, =
2025, at 2:29=E2=80=AFPM, Greg Sabino Mullane &lt;htamfids@gmail.com&gt; =
wrote:<br><div><blockquote type=3D"cite"><br =
class=3D"Apple-interchange-newline"><div><div dir=3D"ltr"><div =
dir=3D"ltr">On Wed, Apr 16, 2025 at 2:18=E2=80=AFAM Ashutosh Bapat =
&lt;<a =
href=3D"mailto:ashutosh.bapat.oss@gmail.com">ashutosh.bapat.oss@gmail.com<=
/a>&gt; wrote:</div><div class=3D"gmail_quote =
gmail_quote_container"><blockquote class=3D"gmail_quote" =
style=3D"margin:0px 0px 0px 0.8ex;border-left:1px solid =
rgb(204,204,204);padding-left:1ex">Users find it a waste of resources to =
deploy 3 big PostgreSQL instances just for HA where 2 suffice even if =
they deploy 3 lightweight DCS instances. Having only some of the nodes =
act as DCS and others purely PostgreSQL nodes will reduce waste of =
resources.<br></blockquote><div><br></div><div>A big problem is that =
putting your DCS into Postgres means your whole system is now =
super-sensitive to IO/WAL-streaming issues, and a busy database doing =
database&nbsp;stuff is going to start affecting the DCS stuff.&nbsp; =
With three lightweight DCS servers, you don't really need to worry about =
how stressed the database servers are. In that way, I feel etcd et al. =
are adhering to the unix philosophy of "do one thing, and do it =
well.=E2=80=9D</div></div></div></div></blockquote></div><div><br></div>=E2=
=80=A6 unless we added multiple WAL streams. That would allow for =
splitting WAL traffic across multiple devices as well as providing =
better support for configurations that don=E2=80=99t replicate the =
entire cluster. The current situation where delayed replication of a =
single table mandates retention of all the WAL for the entire cluster is =
less than ideal.<br><div><br></div></body></html>=

--Apple-Mail=_DFA44C2C-12DB-4491-89D6-2CEE4048254C--