Workshop: Apache Iceberg in Practice: Build, Evolve, and Optimize

Designing high-performance open lakehouse tables is only the starting point. As business requirements pivot, schema definitions shift, and data volumes surge, keeping Apache Iceberg tables performant, reliable, and healthy requires a clear operational framework.

Join Starburst for an interactive, hands-on workshop led by Marius Grama, Engineering Manager at Starburst, titled “Apache Iceberg in Practice: Build, Evolve, and Optimize.” Rather than staying high-level, this session tracks a single Iceberg table across its complete production lifecycle—from initial creation and data hydration to schema evolution and ongoing table maintenance.

By dissecting how Apache Iceberg manages snapshot metadata, manifests, and versioning under the hood, you will walk away with a practical, step-by-step operating model you can implement immediately within your data lakehouse architecture.

What You’ll Learn:

• Table Foundation & Querying: Create, populate, and query Apache Iceberg tables designed for enterprise scale.
• Under-the-Hood Mechanics: Understand exactly how Iceberg stores metadata, tracks snapshots, and versions state changes across distributed storage.
• Seamless Schema & Partition Evolution: Alter column schemas and partition specs on the fly—completely eliminating costly, full-table data rewrites.
• Zero-Copy Branching: Leverage Git-like branching in Iceberg to stage, test, and validate data modifications safely before merging them into production.
• Long-Term Performance Optimization: Implement maintenance strategies (compaction, snapshot expiration, and manifest rewrites) to prevent table degradation as workloads scale.

Whether you are a data engineer, lakehouse architect, or analytics platform administrator, this workshop provides the tactical knowledge needed to build resilient, self-sustaining Iceberg tables that scale effortlessly.