ACID Tables, Time Travel & Production-Grade Lakehouse Operations
Modern data lakes break down the moment data needs to be updated, deleted, audited, or governed.
Apache Iceberg is the technology that fixes this — and it is the foundation of modern Lakehouse architectures.
This course teaches Apache Iceberg from first principles to production operations, specifically for Data Engineers building Lakehouse systems on AWS using Athena and Glue.
You will learn why Iceberg exists, how it works internally, and how it is operated in real production environments — not just how to create tables.
This course is part of the RADE Diamond Membership – Applied Data Engineering Mastery Program and is a core pillar of the Lakehouse Mastery track, building directly on:
Glue Catalog & Crawler Foundations
Athena (ELT & optimization)
This is not a syntax-only course.
You will understand:
Why file formats (Parquet) are insufficient
Why table formats exist
How Iceberg enables ACID, UPDATE, DELETE, MERGE
How Iceberg’s snapshot-based architecture actually works
How metadata, manifests, and data files interact
How Iceberg tables are maintained, optimized, and governed in production
This is the level of understanding expected from Senior Data Engineers and Lakehouse Architects.
By the end of this course, you will be able to:
Explain Iceberg internals confidently in interviews
Design transactional Lakehouse tables
Implement MERGE-based upserts
Use time travel for audit, rollback, and debugging
Apply schema evolution without downtime
Choose correct partitioning strategies
Run production-grade optimization & maintenance workflows
Decide between Copy-on-Write vs Merge-on-Read