PostgreSQL February Meetup in Berlin

Posted by ads on Wednesday, 2024-02-07
Posted in [Berlin][Events][Postgresql-News]

On February 6th, 2024, we had the PostgreSQL February Meetup in Berlin.

Mid of January, Ryan Booz contacted me and told me that he will be in Berlin, right after FOSDEM. And asked if we can organize a Meetup. It’s a bit short notice, but why not. A quick count during the Meetup showed 32 attendees.

PostgreSQL February Meetup in Berlin, at the Redgate office
PostgreSQL February Meetup in Berlin, at the Redgate office

Redgate Software provided us with a meeting room, in their all brand new Berlin office at Mindspace. As well as with pizza and drinks.

We had two speakers:

Ryan Booz

Ryan is using PostgreSQL in his talk to do “Super-powered Data Transformation”. In the examples he loads raw Wordle Tweets into the database, and then shows how PostgreSQL can be used to extract the word rows, gather statistics, and transform the results. For many of the examples he’s using Common Table Expressions and first models a small subset of the data, which then later can be applied to the entire available dataset.

In another set of examples he shows how Advent of Code challenges can be solved entirely in PostgreSQL. That’s not an easy task, I participated in the same challenges.

You can find the slides for the talk on Ryans GitHub site.

Ryan Booz
Ryan Booz

Artur Zakirov

Artur is one of the pg_repack maintainers.

pg_repack is a tool which reorganizes tables in PostgreSQL, with almost no locks. Heavily bloated tables in PostgreSQL usually require a VACUUM FULL in order to get rid of the bloat, which locks the table for the entire time the Vacuum process is running. This is often not acceptable in production environments. Instead pg_repack builds a new table without table bloat, and additionally tracks the changes in the original table. Then the new table is switched in and the bloated table is dropped, which happens in a very short timespan.

Artur explains in his talk how pg_repack works, the different steps which are necessary to build a new table without bloat. He also shows statistics about some of the locking issues in databases which occur in databases with heavy write load.

The slides for Arturs talk are available here.

Artur Zakirov
Artur Zakirov

Social Track

Between the two talks we had time to enjoy pizza, soft drinks and hot beverages. That’s a good opportunity to talk with the other participants.

After the second talk we did stand together for a while, and discussed what we learned, asked questions to the speakers, and also future opportunities for Meetups.

If you like to present your talk at the Meetup, please centact the organizers through the Meetup website.

And if your company wants to shine and host a Meetup (an excellent opportunity for your recruiting team, by the way), please also contact the organizers.

Berlin PostgreSQL Meetup

The PostgreSQL Meetup group in Berlin exists for several years now, and currently has around 1400 members.

Categories: [Berlin] [Events] [Postgresql-News]