Neues vom PostgreSQL Planet

Neues vom PostgreSQL Planet Feed abonnieren
Planet PostgreSQL
Aktualisiert: vor 3 Stunden 1 Minute

Dave Stokes: Loading The Titanic Passenger Data Into PostgreSQL With DBeaver Part 2

22. Oktober 2025 - 14:08

In the last edition of this blog, the passenger list data from the HMS Titanic was loaded into a preliminary database. Now it is time to refine.

I am using DBeaver Enterprise 25.2.0. PostgreSQL 18, and Github Copilot with Gpt-4. 

Prompt: Any recommendations on improving this table for storage efficiency? This prompt was entered into the DBeaver AI Assistant.

Jan Wieremjewicz: Say Hello to OIDC in PostgreSQL 18!

22. Oktober 2025 - 13:00

If you’ve ever wondered how to set up OpenID Connect (OIDC) authentication in PostgreSQL, the wait is almost over.

We’ve spent some time exploring what it would take to make OIDC easier and more reliable to use with PostgreSQL. And now, we’re happy to share the first results of that work.

Why OIDC, and why now?

We’ve spoken to some of our customers and noticed a trend of moving away from LDAP to OIDC. Our MongoDB product is already providing OIDC integration and the team working on PostgreSQL products saw an opportunity coming with PostgreSQL 18.

Hubert 'depesz' Lubaczewski: Waiting for PostgreSQL 19 – Support COPY TO for partitioned tables.

22. Oktober 2025 - 12:05
On 20th of October 2025, Masahiko Sawada committed patch: Support COPY TO for partitioned tables.   Previously, COPY TO command didn't support directly specifying partitioned tables so users had to use COPY (SELECT ...) TO variant.   This commit adds direct COPY TO support for partitioned tables, improving both usability and performance. Performance tests show … Continue reading "Waiting for PostgreSQL 19 – Support COPY TO for partitioned tables."

Chao Li: Understanding the Execution Plan of a Hash Join

22. Oktober 2025 - 10:26

A hash join is one of the most common join methods used by PostgreSQL and other relational databases. It works by building a hash table from the smaller input (called the build side) and then probing it with rows from the larger input (the probe side) to find matching join keys.
Hash joins are especially efficient for large, unsorted datasets—particularly when there are no useful indexes on the join columns.

This post uses a concrete example to explain how a hash join works. The example is run on PostgreSQL 18.

Pavlo Golub: Contributions for week 42, 2025

21. Oktober 2025 - 15:13

Belgium Meetup: Tuesday, 14 October 2025 - organized by Boriss Mejias & Stefan Fercot

Speakers:

  • Niradj Selvam
  • Stefan Fercot

Pavlo Golub: Contributions for week 41, 2025

21. Oktober 2025 - 15:12

Barcelona PostgreSQL User Group met on Tuesday, Oct 7, organized by Dave Pitts & Lauro Ojeda

Speakers - Dave Pitts - Lauro Ojeda

Ellyne Phneah released two books: * The Social Code: Building a PostgreSQL Community Wired for Belonging * Decode PostgreSQL: Understanding the World's Most Powerful Open-Source Database Without Writing Code

Seiten