Neues vom PostgreSQL Planet

     Authors find it challenging to write a comprehensive book about PostgreSQL for several reasons. The first is that a new version of PostgreSQL is released every year with changes, new features, and tweaks.  I am lucky to have been able to review several new database titles each year, and I would like to let you know that Hans-Jurgen Schonig's Mastering PostgreSQL is a well-written reference. 

PostgreSQL has a fairly efficient buffer manager that handles data movement between shared memory and disk storage. Among other things, its primary purpose is to serve as a disk cache for the client backend processes for quick access to data. The size of this buffer is dictated by the setting shared_buffers in postgresql.conf, and its default value is 128 MB.

Broadcasting a quick tip from a real-life “tasklet” where a customer was genuinely surprised why a seemingly logical fix didn’t show up on Google / LLM-s. So let’s see if we can improve on that situation. So in short a large central table was getting too bloated (fragmented) and query...

PostgreSQL 18 is on the way, bringing a set of improvements that many organizations will find useful. It’s not a revolutionary release, but it does move things in a good direction, especially in performance, replication, and simplifying daily operations. For teams already using PostgreSQL, it’s a good time to look into what’s new. For others […]

PostgreSQL Person of the Week Interview with Shlok Kumar Kyal: My name is Shlok Kumar Kyal. I live in Bangalore, India. I am currently part of the PostgreSQL Open Source team at Fujitsu.

I often hear that PostgreSQL is not suitable for solving analytics problems, referencing TPC-H or ClickBench results as evidence. Surely, handling a straightforward task like sorting through 100 million rows on disk and calculating a set of aggregates, you would get stuck on the storage format and parallelisation issues that limit the ability to optimise the DBMS.

On 29th of July 2025, David Rowley committed patch: Display Memoize planner estimates in EXPLAIN   There've been a few complaints that it can be overly difficult to figure out why the planner picked a Memoize plan. To help address that, here we adjust the EXPLAIN output to display the following additional details:   1) … Continue reading "Waiting for PostgreSQL 19 – Display Memoize planner estimates in EXPLAIN"

Just a short post, because I thought this was pretty remarkable. Below, I have screenshots showing the CPU utilization of two AWS instances in us-west-2 which are running an identical workload.

I had the pleasure of speaking at PGConf.dev in Montreal, where I delivered a session on Columnar Data Store—a topic that’s quietly transforming the way we handle large-scale analytics in PostgreSQL. In this post, I’ll walk you through the core ideas from my talk and explain why columnar storage isn’t just a performance boost—it’s a game changer for modern data workloads.

Catching with all contributions. PGDay UK published their schedule. Registration is still open. Here is their program committee

PgDay UK

Sharing my progress on how I am working on ABI compliance reporting for PostgreSQL as a BuildFarm Module. In this post, I walk through the new workflow, highlighting lessons learned from community feedback, and discuss upcoming improvements. If you're curious about ABI checking for Postgres Binaries and how it integrates with BuildFarm, this post is for you!

You probably saw the story about your public ChatGPT queries being indexed by Google and other search engines. Yikes! 
You may want AI with your data but do not wish your prompts to end up in a search engine, especially if you are working with sensitive data.  

Nordic PGDay 2026 will return to Helsinki, Finland, on March 24. See our website for details, and stay tuned for more!

This week's edition is unfortunately delayed due to personal reasons.

PostgreSQL 19 changes this week psql : variable SERVICEFILE added PostgreSQL 18 changes this week

Main item of note this week: PostgreSQL 18 Beta 2 has been released

more...

It's Sunday and I'm now sitting in the train from Brest to Paris where I will be changing to Germany, on the way back from the annual Debian conference. A full week of presentations, discussions, talks and socializing is laying behind me and my head is still spinning from the intensity.

Pollito and the gang of DebConf mascots wearing their conference badges (photo: Christoph Berg)

Table of Contents

Ollama is an open-source project that simplifies running large language models (LLMs). It runs locally on your machine, and you can choose from multiple LLMs. Keeping all the data on your machine should provide a security bonus.   

DBeaver is a universal database tool with an AI Assistant. This assistant provides an extra layer of security by allowing you to lock down what is shared, with the default being metadata only, no data.

PostgreSQL Person of the Week Interview with Polina Bungina: I am Polina, a PostgreSQL database engineer at Zalando, located in Berlin, Germany. However I am originally from St. Petersburg, Russia.

A case study on how Bilt reduced Postgres query timeouts by 95% with ParadeDB

Greetings!

I work in Tantor Labs as a database developer and naturally I am fond of databases. Once during reading the red book I have decided to study planner deeply. Main part of relational database planner is join ordering and I came across DPhyp algorithm that is used in most modern (and not so much) databases. I wonder - is there is anything in PostgreSQL? Surprisingly, nothing. Well, if something does not exist, you need to create it yourself.

I'm excited to announce that Postgres AI has started work on a new project – open-source Self-Driving Postgres (SDP).

In the AI era, Postgres is the natural choice for AI builders. With fast-growing database clusters, the highest level of automation is essential. AI-driven growth demands efficient, proactive, and intelligent database management. Our goal is to reduce manual interventions as much as possible to achieve the highest level of operational efficiency and reliability.

How can we define levels of automation?