Content-Type: text/plain;
	charset=us-ascii
Mime-Version: 1.0 (Mac OS X Mail 16.0 \(3776.700.51\))
Subject: Re: Loading data from one table to another
From: Scott Ribe <scott_ribe@elevated-dev.com>
In-Reply-To: 
 <DB7PR03MB378887A0840FDAECF62410CD8C7B2@DB7PR03MB3788.eurprd03.prod.outlook.com>
Date: Sun, 13 Oct 2024 08:34:51 -0600
Cc: "pgsql-admin@lists.postgresql.org" <pgsql-admin@lists.postgresql.org>
Content-Transfer-Encoding: quoted-printable
Message-Id: <BD7CC9C3-75C7-4D7C-A5F0-09ACC9828806@elevated-dev.com>
References: 
 <DB7PR03MB378887A0840FDAECF62410CD8C7B2@DB7PR03MB3788.eurprd03.prod.outlook.com>
To: Sabyasachi Mukherjee <mukherjee.sabyasachi@outlook.com>
Archived-At: 
 <https://www.postgresql.org/message-id/BD7CC9C3-75C7-4D7C-A5F0-09ACC9828806%40elevated-dev.com>
Precedence: bulk

Failing to allocate a memory segment of only 1MB seems suspicious. What =
are your hardware specs?

You might want to look at your memory & parallel query settings in =
postgresql.conf--perhaps some of those are set too high relative to your =
server's memory.

I wouldn't really expect queries to be slow on 99GB, by the way. That is =
not a large database. Partitioning is reasonable if you want to prune =
old data periodically, as then you can just drop old partitions. But =
unless your queries are regularly have to perform table scans, I =
wouldn't expect them to be slow. And if they are performing table scans, =
you'd need to investigate how to fix that--what indexes to add. =
Partitioning will get you incremental performance gains in such a case, =
while the right index will get you exponential gains.=