Apache Iceberg™ Europe Community Meetup - Warsaw
Apache Iceberg™ Europe Meetup - Warsaw
Join us for the Apache Iceberg™ Europe Meetup! Our event is hosted in Warsaw co-hosted by Google Cloud, Starburst and Vakamo.
🎟️ When registering, please select one of the two ticket types:
In-Person Ticket: Join us on-site! Your name will be used to pre-register for venue access.
Remote-Only Ticket: Can’t make it in person? No worries—register to join the live stream, receive event recordings, and stay connected with the community.
Livestream
Agenda
5:00 pm – Registration & Networking
6:00 pm – Meetup starts
🎙️ Viktor Kessler - Power of Apache Iceberg Catalog REST Specification
🎙️ Piotr Findeisen and Piotr Rżysko - GPU-enabled Iceberg Query Acceleration
7:00 pm – 7:30 pm Networking break
🎙️ Vladislav Sidorovich - Multi-cloud data access with Gemini and Apache Iceberg
🎙️ Przemysław Hejman - Pick Your Warehouse: Iceberg as Neutral Ground for AI Agent Trajectory Data
How to Get to the Venue
Address:
Rondo Daszyńskiego 2C, 00-843 Warszawa, Poland
Building Access
Goolge Cloud
🪪 ID Requirements
Presentations & Speakers
🌟 Power of Apache Iceberg Catalog REST Specification
Abstract
The Iceberg REST Catalog specification began as a way to stop every engine from shipping its own catalog client. It has become the control plane of the open lakehouse: one API that Spark, Trino, Flink, DuckDB, PyIceberg and many others use to find tables, commit changes, get access credentials and enforce governance.
In this session we look at how the spec has grown and where it's heading. We start with the basics: namespaces, tables and atomic commits. Then we cover the features that turned it into a real platform API:
views
multi-table transactions
credential vending
server-side scan planning
the catalog as the place where access control and auditing happen
Next we ask what comes after Iceberg tables. Data and AI work runs on more than tables. It also uses documents, images, audio, model files, embeddings and notebooks, and most of that sits ungoverned in object storage. We show how the same REST approach (the same endpoints, auth, permissions and credential vending) can be extended to generic tables, views and file-based datasets. That brings all of the lake, unstructured data included, under one catalog.
To make this concrete, the live demo uses Lakekeeper and jupyterlab-lakekeeper, an open-source JupyterLab extension:
browse warehouses, namespaces and tables from a sidebar, signed in with your own identity (JupyterHub, Keycloak or service credentials)
work with unstructured datasets like normal folders: open, upload, drag and drop, and save notebooks straight into governed storage
preview and query Iceberg tables with SQL, and reach the same data from a notebook with
connect()andsql()connect to several catalogs from one place, with nothing more than a URL
If you work on data platforms or AI infrastructure, you'll leave with a clear picture of why the REST catalog matters more than the table format beneath it, and how to use it to govern all of your lake data.
Viktor Kessler, is Co-Founder of Vakamo and the creator of Lakekeeper, an Apache Licensed Iceberg REST Catalog. He’s a big believer in open standards like Apache Iceberg, which he sees as the backbone of today’s modern, composable Data & Analytics systems.
🌟 GPU-enabled Iceberg Query Acceleration
In this talk, we will delve into how Starburst has enhanced its query engine to accelerate queries using Nvidia GPUs over Iceberg data. We will share insights into the query engine and processing modifications made and our partnership with Nvidia, along with some benchmark results.
Piotr Findeisen is a co-founder and Distinguished Software Engineer at Starburst. He has been a Trino maintainer since 2017, back when it was still called Presto. His Trino work spans the whole project, from connectors to client tools, and includes the cost-based optimizer, work on the execution engine and SPI, sophisticated connector pushdowns (predicates, joins, aggregations), and, recently, the high-precision NUMBER type. He is also a committer on Apache Iceberg and Apache DataFusion. In Iceberg, he created the Puffin file format for table statistics and indexes, and built the Trino support for it. In DataFusion, he worked on SQL correctness and API design to make it a great building block for data processing applications such as the Rust-based dbt Fusion engine. He is currently building GPU-accelerated query execution for Starburst's Trino-based engine on top of NVIDIA RAPIDS cuDF.
🌟 Multi-cloud data access with Gemini and Apache Iceberg
Discover how to unlock multi-engine analytics by running native BigQuery SQL (DML, DDL, and AI functions) directly over Apache Iceberg tables — and explore the architecture behind it.
We'll walk the architecture end to end: how the Lakehouse Iceberg REST Catalog keeps metadata transactionally consistent while your data stays in open Iceberg format in your own Cloud Storage buckets, and how query engines read and write the same tables without vendor lock-in.
Vladislav Sidorovich, SWE at Google Cloud