CoursesChallenges

DuckDB Hands-On

Meet a real query engine with the cluster removed. Every concept from Query Engine Foundations becomes visible in a single process you can attach a debugger to: the vectorized engine up close, reading Parquet/Iceberg/Delta in place, Arrow zero-copy hand-off, the single-writer concurrency model, Quack making DuckDB plural over a wire, the honest 'when one process beats a cluster' rule, real production use cases, and a capstone that ports a small Spark job.

Intermediate14 chapters· 4h 7m· in Query Engines & OLAP

Course content

  1. 01The Embedded Database PatternFree
  2. 02DuckDB Architecture🔒
  3. 03The Vectorized Engine Up Close🔒
  4. 04SQL Surface🔒
  5. 05Reading Parquet & CSV Directly🔒
  6. 06Reading Iceberg & Delta🔒
  7. 07Arrow Integration🔒
  8. 08Persistent Databases🔒
  9. 09Extensions🔒
  10. 10The Concurrency Model🔒
  11. 11DuckDB Quack: Making the Duck Plural🔒
  12. 12When DuckDB Beats a Cluster🔒
  13. 13Real Practical Use Cases of DuckDB🔒
  14. 14Capstone: Replace a Small Spark Job with DuckDB🔒

Prerequisites

↗Query Engine Foundations

Read the first chapter free

Start reading now — no account required for the free chapters.

Start: The Embedded Database Pattern →More in Query Engines & OLAP