FOSDEM 2026 took place on 31 January and 1 February in Brussels, and it was a great weekend for the ClickHouse community, both inside the conference rooms and out.

## Community Dinner

The weekend kicked off with the ClickHouse community dinner at L'Ultime Atome in Ixelles, Brussels. 106 people joined us for dinner, drinks, and some of the best conversations of the conference. ClickHouse founder Alexey was there to chat and answer questions. Thanks to everyone who came out; it was a fantastic evening (some pictures are at the end of this post).

But it wasn’t just a chance to eat and drink; we had multiple presentations at the event itself.

## Our Talks

### Planes, Ships, Birds – Building Real-Time Visualizations with ClickHouse

**Track:** Geospatial · **Day:** Saturday
**Speaker:** Alexey Milovidov · [Slides](https://presentations.clickhouse.com/2026-fosdem-geospatial/)

This session walked through creating an analytical application from the ground up, covering the entire workflow: data collection, processing, and visualization. The stack uses Leaflet, ClickHouse, and a collection of shell scripts. The finished result is published at [adsb.exposed](https://adsb.exposed/), and the talk uncovered how it was built.

<iframe width="560" height="315" src="https://www.youtube.com/embed/YOlEcBLDVvk?si=5EVmFmI2v--q6DKX" title="YouTube video player" frameborder="0" allow="accelerometer; autoplay; clipboard-write; encrypted-media; gyroscope; picture-in-picture; web-share" referrerpolicy="strict-origin-when-cross-origin" allowfullscreen></iframe>

---

### Hotpatching ClickHouse in Production with XRay

**Track:** LLVM · **Day:** Saturday
**Speaker:** Pablo Marcos · [Slides](https://fosdem.org/2026/events/attachments/N7MVZT-hotpatching-clickhouse-with-llvm-xray/slides/266966/hotpatchi_f1nbs5s.pdf)

Ever been debugging a production issue and wished you'd added just one more log statement — only to face a rebuild, a wait for CI, and a full redeploy? The ClickHouse team integrated LLVM's XRay to solve exactly this. It lets you hot-patch a running production system to inject logging, profiling, and even deliberate delays into any function, with no rebuild required.

XRay reserves space at function entry and exit points that can be atomically patched with custom handlers at runtime. The team built three handler types: LOG to add the trace points you forgot, SLEEP to reproduce or prevent timing-sensitive bugs, and PROFILE for deterministic profiling to complement the existing sampling profiler. The performance overhead when inactive is negligible.

Control is via a SQL query — `SYSTEM INSTRUMENT ADD LOG 'QueryMetricLog::startQuery' 'message'` — which patches the function instantly, with results appearing in `system.trace_log`. The talk also covered the integration challenges (ELF parsing, thread-safety, atomic patching), performance numbers (4–7% binary size increase, near-zero runtime cost), and real production war stories.

<iframe width="560" height="315" src="https://www.youtube.com/embed/DZeVcRD0rWM?si=X47mRrsF1LXTIjLp" title="YouTube video player" frameborder="0" allow="accelerometer; autoplay; clipboard-write; encrypted-media; gyroscope; picture-in-picture; web-share" referrerpolicy="strict-origin-when-cross-origin" allowfullscreen></iframe>

---

### Inverted database indexes: The why, the what, and the how.

**Track:** Databbases · **Day:** Saturday 
**Speaker:** Elmi Ahmadov · [Slides](https://presentations.clickhouse.com/2026-fosdem-inverted-index/Inverted_indexes_the_what_the_why_the_how.pdf)

Database usage in practice often involves heavy text processing. For example, in "observability" use cases, databases must extract, store, and search billions of log messages daily. Most databases, including many column-oriented OLAP databases, struggle with such massive amounts of text data. The only way to process text data at scale is by using specialized inverted indexes in databases.

This presentation explains how inverted indexes work and which (text) search patterns they support. Where appropriate, we describe our experience and the gotchas we encountered when adding an inverted index to ClickHouse, one of the most popular open-source databases for analytics.

<iframe width="560" height="315" src="https://www.youtube.com/embed/6eM1rCZy7XI?si=MwfUq1E_fFmQC6K8" title="YouTube video player" frameborder="0" allow="accelerometer; autoplay; clipboard-write; encrypted-media; gyroscope; picture-in-picture; web-share" referrerpolicy="strict-origin-when-cross-origin" allowfullscreen></iframe>

---

### ClickHouse's C++ and Rust Journey

**Track:** Rust · **Day:** Sunday
**Speaker:** Alexey Milovidov · [Slides](https://presentations.clickhouse.com/2025-p99/)

For a large C++ codebase, a full rewrite to Rust is not an option — only gradual integration with Rust libraries is realistic, and even then there are many complications and rough edges. This talk described the experience of integrating Rust and C++ code in ClickHouse, and some of the weird and unusual problems the team had to overcome along the way.

<iframe width="560" height="315" src="https://www.youtube.com/embed/433zALc6u3Y?si=Ux9Htf42tdWYQas-" title="YouTube video player" frameborder="0" allow="accelerometer; autoplay; clipboard-write; encrypted-media; gyroscope; picture-in-picture; web-share" referrerpolicy="strict-origin-when-cross-origin" allowfullscreen></iframe>

---

### How to Move Bytes Around

**Track:** Software Performance · **Day:** Sunday
**Speaker:** Alexey Milovidov · [Slides](https://presentations.clickhouse.com/2026-fosdem-memcpy/)

If you take a random program and start profiling it, you'll usually find `memcpy` near the top — though that doesn't necessarily mean memcpy is slow. As the abstract puts it: the most hopeless thing a C++/Rust developer can do (while no one is watching) is optimize memcpy to move bytes faster. That's exactly what this talk did.

<iframe width="560" height="315" src="https://www.youtube.com/embed/rPX-3UrFiF4?si=OP7_nRl8v16xccCf" title="YouTube video player" frameborder="0" allow="accelerometer; autoplay; clipboard-write; encrypted-media; gyroscope; picture-in-picture; web-share" referrerpolicy="strict-origin-when-cross-origin" allowfullscreen></iframe>

---

### Contributing to MariaDB & Postgres

**Track:** Databases · **Day:** Saturday
**Speakers:** Georgi Kodinov, Kevin Biju

A two-part talk on contributing to major open-source database projects.

Georgi Kodinov walked through what it actually takes to get a contribution into the MariaDB server codebase, following a real bug fix — two lines of code — through its entire review process, and explaining why the bar is higher than it might seem on a project of this scale.

Kevin Biju traced his own path from setting up a local Postgres build to contributing bug-fix patches and documentation updates. The talk outlines how the Postgres development process and community operate, with the aim of demystifying the process so more engineers feel confident making their first (or next) patch.

<iframe width="560" height="315" src="https://www.youtube.com/embed/BgODDRqp6yA?si=j2NchIseWMoIWAvW" title="YouTube video player" frameborder="0" allow="accelerometer; autoplay; clipboard-write; encrypted-media; gyroscope; picture-in-picture; web-share" referrerpolicy="strict-origin-when-cross-origin" allowfullscreen></iframe>

---

## From the Community

We weren't the only ones talking about ClickHouse at FOSDEM. Here are some talks from the community that are worth checking out:

### Dynamic Bot Blocking with Web-Server Access-Log Analytics

Alexander Krizhanovsky - [Video](https://fosdem.org/2026/schedule/event/3BMPMU-dynamic_bot_blocking_with_web-server_access-log_analytics/)

Bots generate roughly half of all Internet traffic. Some are clearly malicious (password crackers, vulnerability scanners, application-level/L7 DDoS), and others are merely unwanted (web scrapers, carting, appointment etc) bots. Traditional challenges (CAPTCHAs, JavaScript checks) degrade user experience, and some vendors are deprecating them. An alternative is traffic and behavior analytics, which is much more sophisticated, but can be far more effective.

Complicating matters, there are cloud services not only helping to bypass challenges, but also mimic browsers and human behavior. It's tough to build a solid protection system withstand such proxy services.

In this talk, we present WebShield, a small open-source Python daemon that analyzes Tempesta FW, an open-source web accelerator, access logs and dynamically classifies and blocks bad bots.

You'll learn: * Which bots are easy to detect (e.g., L7 DDoS, password crackers) and which are harder (e.g., scrapers, carting/checkout abuse). * Why your secret weapon is your users’ access patterns and traffic statistics—and how to use them. * How to efficiently deliver web-server access logs to an analytics database (e.g., ClickHouse). * Traffic fingerprints (JA3, JA4, p0f): how they’re computed and their applicability for machine learning * Tempesta Fingerprints: lightweight fingerprints designed for automatic web clients clustering. * How to correlate multiple traffic characteristics and catch lazy bot developers. * Baseline models for access-log analytics and how to validate them. * How to block large botnets without blocking half the Internet. * Scoring, behavioral analysis, and other advanced techniques are not yet implemented

## Photos

We wish you could have been there. Here are some images that capture a sense of the experience. Do you want to join the next one? Let us know and we can help with your presentation.

large_IMG_2418.jpg

medium_IMG_2418.jpg

small_IMG_2418.jpg

thumbnail_IMG_2418.jpg

IMG_2418.jpg

large_PXL_20260201_105432126.jpg

medium_PXL_20260201_105432126.jpg

small_PXL_20260201_105432126.jpg

thumbnail_PXL_20260201_105432126.jpg

PXL_20260201_105432126.jpg

large_1000137544.jpg

medium_1000137544.jpg

small_1000137544.jpg

thumbnail_1000137544.jpg

1000137544.jpg

large_PXL_20260131_194611228.jpg

medium_PXL_20260131_194611228.jpg

small_PXL_20260131_194611228.jpg

thumbnail_PXL_20260131_194611228.jpg

PXL_20260131_194611228.jpg

large_PXL_20260201_141128012.jpg

medium_PXL_20260201_141128012.jpg

small_PXL_20260201_141128012.jpg

thumbnail_PXL_20260201_141128012.jpg

PXL_20260201_141128012.jpg

large_PXL_20260131_194552756.jpg

medium_PXL_20260131_194552756.jpg

small_PXL_20260131_194552756.jpg

thumbnail_PXL_20260131_194552756.jpg

PXL_20260131_194552756.jpg

FOSDEM 2026 took place on 31 January and 1 February in Brussels, and it was a great weekend for the ClickHouse community, both inside the conference rooms and out.

large_Fosdem26.png

medium_Fosdem26.png

small_Fosdem26.png

thumbnail_Fosdem26.png

Fosdem26.png

ClickHouse at FOSDEM 2026

This year, ClickHouse made a strong showing at FOSDEM 2025, with several team members traveling to Belgium and multiple talks covering everything from powerful new JSON data types to the challenges of fuzzing databases. If you weren’t able to attend (or just want a recap), here’s what went down in Brussels.

## What is FOSDEM?

For those unfamiliar, **FOSDEM** (Free and Open Source Developers' European Meeting) is one of the biggest open-source conferences in the world. Every year, thousands of developers, engineers, and open-source enthusiasts gather in Brussels to share knowledge, showcase projects, and engage in deep technical discussions. It's a place where open-source software meets innovation, and where deep-dive technical talks are the norm.

And this year, ClickHouse was **well represented**.

## The ClickHouse Talks

### How We Built a New Powerful JSON Data Type for ClickHouse

JSON support is a hot topic for ClickHouse users, and at FOSDEM, we shared details about the **new JSON data type** we built. This talk covered **why** we introduced it, how it **differs from traditional JSON handling**, and what kind of performance benefits users can expect. If you're working with semi-structured data at scale, this was a must-watch.

**Speaker:** Robert Schulze

**Abstract:** JSON has become the lingua franca for handling semi-structured and unstructured data in modern data systems. Whether it’s in logging and observability scenarios, real-time data streaming, mobile app storage, or machine learning pipelines, JSON’s flexible structure makes it the go-to format for capturing and transmitting data across distributed systems.

At ClickHouse, we’ve long recognized the importance of seamless JSON support. But as simple as JSON seems, leveraging it effectively at scale presents unique challenges. In this talk we will discuss how we built a new powerful JSON data type for ClickHouse with true column-oriented storage, support for dynamically changing data structure and ability to query individual JSON paths really fast.

Links related to the topic:

* [RFC: Semistructured Columns #54864](https://github.com/ClickHouse/ClickHouse/issues/54864)
* [Implement Variant data type #58047](https://github.com/ClickHouse/ClickHouse/pull/58047)
* [Implement Dynamic data type #63058](https://github.com/ClickHouse/ClickHouse/pull/63058)
* [Implement new JSON data type. #66444](https://github.com/ClickHouse/ClickHouse/pull/66444)

<iframe width="768" height="432" src="https://www.youtube.com/embed/JohjugQedSk?si=6jBEdh9MHSoAP7yA" title="YouTube video player" frameborder="0" allow="accelerometer; autoplay; clipboard-write; encrypted-media; gyroscope; picture-in-picture; web-share" referrerpolicy="strict-origin-when-cross-origin" allowfullscreen></iframe>

### Fuzzing Databases is Difficult

Database reliability and security are always important priorities, and fuzz testing is one way to uncover hidden bugs. But as it turns out, **fuzzing databases is hard**. This talk dove into the complexities of fuzz testing, why traditional approaches don't always work well for databases, and what strategies can improve effectiveness. If you’re interested in database internals or security, this one was for you.

**Speaker:** Pedro Ferreira 

**Abstract:** After fuzzing databases for the last 3 years, I learned that simple design decisions on a fuzzer impact on the issues it can ever find. In this talk I would to address some of those decisions. As an example, I would to discuss about the design of BuzzHouse, a new database fuzzer to test ClickHouse.

Links related to the topic: 
[First iteration of Buzzhouse #71085
](https://github.com/ClickHouse/ClickHouse/pull/71085)

<iframe width="768" height="432" src="https://www.youtube.com/embed/CW4Ntdtp7lg?si=8zlyBIg8xzXiecCF" title="YouTube video player" frameborder="0" allow="accelerometer; autoplay; clipboard-write; encrypted-media; gyroscope; picture-in-picture; web-share" referrerpolicy="strict-origin-when-cross-origin" allowfullscreen></iframe>

### ** rDNS Map: The Internet in Your Hands**

Reverse DNS (rDNS) is an often-overlooked but powerful tool for understanding the structure of the internet. This talk explored how we can **map out rDNS data** at scale and why it's useful for network monitoring, security research, and even performance optimization.

**Speaker:** Alexey Milovidov

**Abstract:** I've created an rDNS map, available at https://reversedns.space/, and I want to tell you how. It was not hard to do, but there was a lot of unusual and amusing stuff in the process.

<iframe width="768" height="432" src="https://www.youtube.com/embed/0hDOr9Pp1-4?si=ppV4bjsQUqxwz6b8" title="YouTube video player" frameborder="0" allow="accelerometer; autoplay; clipboard-write; encrypted-media; gyroscope; picture-in-picture; web-share" referrerpolicy="strict-origin-when-cross-origin" allowfullscreen></iframe>

## Wrapping Up with a Community Dinner

Of course, no ClickHouse event is complete without some time with the community. After a full day of talks, we gathered with our community members for dinner and drinks in Brussels. It was a chance to meet users, discuss database internals, and share ideas over great food and drinks. If you were there, thanks for joining us! If not, we hope to see you at the next one. 

In the meantime, feel free to join us on [Slack](https://clickhouse.com/slack) or give the [ClickHouse repository](https://github.com/ClickHouse/clickhouse) a star.

## Some Photos

![FOSDEM_Robert.png](https://cms.clickhouse-dev.com:1337/uploads/FOSDEM_Robert_7ae18824d8.png)

![FOSDEM_Pedro.png](https://cms.clickhouse-dev.com:1337/uploads/FOSDEM_Pedro_281eabed62.png)

![FOSDEM_Alexey.png](https://cms.clickhouse-dev.com:1337/uploads/FOSDEM_Alexey_f71f4433fb.png)

![FOSDEM_Room_Full.png](https://cms.clickhouse-dev.com:1337/uploads/FOSDEM_Room_Full_42d27aab9b.png)

![FOSDEM_Dinner_Venue.jpg](https://cms.clickhouse-dev.com:1337/uploads/FOSDEM_Dinner_Venue_2d60dbbefb.jpg)

![FOSDEM_Dinner_Full.jpg](https://cms.clickhouse-dev.com:1337/uploads/FOSDEM_Dinner_Full_bbae49afa6.jpg)


This year, ClickHouse made a strong showing at FOSDEM 2025, with several team members traveling to Belgium and multiple talks covering everything from powerful new JSON data types to the challenges of fuzzing databases.

large_ClickHouse_FOSDEM.png

medium_ClickHouse_FOSDEM.png

small_ClickHouse_FOSDEM.png

thumbnail_ClickHouse_FOSDEM.png

ClickHouse_FOSDEM.png

ClickHouse at FOSDEM 2025: talks, tech, and a community dinner

Today we are pleased to announce the availability of an integration allowing ClickHouse users to quickly, and simply, install and use Deepnote.


![deepnote-2.webp](/uploads/deepnote_2_3a74fdc5db.webp)

If you understand the impact already, then get started with the [documentation](https://clickhouse.com/docs/en/integrations/deepnote/). 

Our friends at Deepnote have also written an excellent, detailed, [blog post](https://deepnote.com/blog/clickhouse-cl4zs29ikocet0blrv08ishlj) on the topic.


## **What is Deepnote?**

So, what is it?

[Deepnote](https://www.deepnote.com/) is a collaborative data notebook for teams to discover and share insights. In addition to being Jupyter-compatible, it works in the cloud and provides one central place to collaborate and work on data science projects efficiently.

Deepnote provides first-class support for SQL. This means you get to query your ClickHouse database right from your notebook. Transitions between Python and SQL are seamless as there’s no need for a Python connector. With Deepnote, you get all the bells and whistles of a SQL editor right in your notebook, including formatting, autocomplete, and linting.


## **Getting Started**

If you want to explore an interactive example of querying ClickHouse from Deepnote data notebooks, click the button below to launch a template project connected to the ClickHouse playground.

[![launch-in-deepnote.svg](/uploads/launch_in_deepnote_b9ad65cb49.svg)](https://deepnote.com/launch?template=ClickHouse%20and%20Deepnotenk)

## **Sample Queries and a Challenge**

If you don’t have a dataset handy, the ClickHouse Playground is convenient for exploring ClickHouse using Deepnote. While multiple datasets are available, it is worth issuing a bit of a challenge of exploration.

First, look at the [GitHub Archive](https://ghe.clickhouse.tech/) data that we persist in Playground.

Second, explore the data after connecting your Deepnote notebook to the playground (sample queries are available at the link above).

Finally, [Tweet](https://twitter.com/clickhousedb), [Telegram](https://telegram.me/clickhouse_en), [Slack](https://join.slack.com/t/clickhousedb/shared_invite/zt-rxm3rdrk-lIUmhLC3V8WTaL0TGxsOmg), or [contact us](https://clickhouse.com/contact/) to let us know what you have created.

The best submission will receive a ClickHouse t-shirt!


## **Summary**

ClickHouse is the fastest OLAP database on the planet. But data is only as compelling, and valuable, as the insight you derive from it. We are delighted to partner to provide a new, exciting, and collaborative way of interacting with your data.


Today we are pleased to announce the availability of an integration allowing ClickHouse users to quickly, and simply, install and use Deepnote.

large_DeepnoteAndClickHouse.webp

medium_DeepnoteAndClickHouse.webp

small_DeepnoteAndClickHouse.webp

thumbnail_DeepnoteAndClickHouse.webp

DeepnoteAndClickHouse.webp

ClickHouse + Deepnote: Data Notebooks & Collaborative Analytics

## We're meeting in person! Join us on March 9th 2023 in Amsterdam. [Register here](https://www.meetup.com/clickhouse-netherlands-user-group/events/291485868/).





**tl;dr – Join us virtually or June 8th in Amsterdam – [Watch Recording](https://www.youtube.com/watch?v=DP7l6Swkskw&ab_channel=ClickHouse)!**

Following the success of our day of virtual meetups, we’ve decided it is time to meet again…in person.

A significant portion of the ClickHouse team will be in Amsterdam. So we decided to take the opportunity to get together, talk tech, and share some food. Please [register now](https://www.meetup.com/clickhouse-netherlands-user-group/events/286017044/) so that we can ensure sufficient space and food. It’s not just an opportunity to network, though. You will hear stories of ClickHouse powering a performance & monitoring service and a detailed exploration of how, and why, ClickHouse is so fast.

We’ve identified a venue…

We’ve found speakers…

Our founding team will be in town…

All that is missing is you!


## **Why a Meetup?**

ClickHouse, as an open-source company, is rooted deeply in the importance of community. The opportunity to interact with our users, in physical space, is core to the identity of our team.

In addition, our history (as a company and individuals) is built on the power of conversation. Some of the stories we share, and memories we cherish, involve a small venue, lightning talks, and conversations that last until security reminds us that the booking period is over.

Our community matters. Our interaction with our community is core to who we are and who we aspire to be.


## **Agenda**

The current agenda is as follows:

17:45 Doors Open – Networking & Snacks

18:00 Welcome

18:05 Lorenzo Mangani, QXIP BV – 45 mins

18:50 Lightning Talks! (Will it be you???)

19:00 Coffee Break

19:15 Alexey Milovidov, ClickHouse – 45 mins

20:00 Beverages, Food, and Conversation

![pasted image 0.png](/uploads/pasted_image_0_33b308b535.png)

We will be gathering at the beautiful, and unique, Night venue in Venture Studio at the [Startup Village](https://www.startupvillage.nl/). A big thanks to the team there and Rabobank for hosting us.

If you have an idea for a Lightning Talk please let us know (tyler@clickhouse.com). We are finalizing them as we speak.


## **Speaker Bios:**

**Lorenzo Mangani** – CEO of Amsterdam based [QXIP BV](https://qxip.net/), Co-Founder of Uptrace & cLoki, directing a dream team specialized in active and passive large scale monitoring and ingestion of Traces, Logs, Packets and Metrics on top of ClickHouse

**Alexey Milovidov**, CTO & Co-Founder ClickHouse: Alexey Milovidov is the original creator, co-founder and CTO of ClickHouse. He started work on ClickHouse in 2009 and initiated its release to open-source in 2016.

It’s not just an opportunity to network, though. You will hear stories of ClickHouse powering a performance & monitoring service and a detailed exploration of how, and why, ClickHouse is so fast.

Tyler Hannan

Blog posts by Tyler Hannan

ClickHouse at FOSDEM 2026

ClickHouse at FOSDEM 2025: talks, tech, and a community dinner

ClickHouse + Deepnote: Data Notebooks & Collaborative Analytics

Amsterdam Meetup With The ClickHouse Team – June 8th, 2022