Horton

Re: [Proposal] Fully WAL logged CREATE DATABASE - No Checkpoints

Robert Haas <robertmhaas@gmail.com>

View thread View on postgr.es
From: Robert Haas <robertmhaas@gmail.com>
To: Dilip Kumar <dilipbalaut@gmail.com>
Cc: Alvaro Herrera <alvherre@alvh.no-ip.org>, Andrew Dunstan <andrew@dunslane.net>, Heikki Linnakangas <hlinnaka@iki.fi>, PostgreSQL Hackers <pgsql-hackers@lists.postgresql.org>
Date: 2022-02-14T13:53:32Z
Lists: pgsql-hackers 
On Mon, Feb 14, 2022 at 12:01 AM Dilip Kumar <dilipbalaut@gmail.com> wrote:
> Well this is not really surprising to me because what I have noticed
> is that with the new approach the createdb time is completely
> dependent upon the template db size.  So if the source db size is 10GB
> it is taking around 80sec and the shared buffers size does not have a
> major impact.  Maybe a very small shared buffer can have more impact
> so I will test that as well.

OK. Well, then this approach is somewhat worse than I expected for
moderately large template databases. But it seems very good for small
template databases, especially when there is other work in progress on
the system.

-- 
Robert Haas
EDB: http://www.enterprisedb.com

Commits

  1. When using the WAL-logged CREATE DATABASE strategy, bulk extend.

  2. Avoid using a fake relcache entry to own an SmgrRelation.

  3. Fix data-corruption hazard in WAL-logged CREATE DATABASE.

  4. initdb: When running CREATE DATABASE, use STRATEGY = WAL_COPY.

  5. Simplify a needlessly-complicated regular expression.

  6. In 020_createdb.pl, change order of command-line arguments.

  7. Add new block-by-block strategy for CREATE DATABASE.

  8. Fix replay of create database records on standby

  9. Refactor code for reading and writing relation map files.

  10. Replace RelationOpenSmgr() with RelationGetSmgr().

  11. Refactor the fsync queue for wider use.