DuckDB in Action

by Mark Needham, Michael Hunger, Michael Simons

Dive into DuckDB and start processing gigabytes of data with ease—all with no data warehouse.

You don’t need expensive hardware or to spin up a whole new cluster whenever you want to analyze a big data set. You just need DuckDB! This modern and fast embedded database runs on a laptop, and lets you easily process data from almost any source, including JSON, CSV, Parquet, SQLite and Postgres. In DuckDB in Action you’ll learn everything you need to know to get the most out of this awesome tool, keep your data secure on prem, and save you hundreds on your cloud bill.

Open up DuckDB in Action and learn how to:

- Read and process data from CSV, JSON and Parquet sources both locally and remote
- Write analytical SQL queries, including aggregations, common table expressions, window functions, special types of joins, and pivot tables
- Use DuckDB from Python, both with SQL and its "Relational"-API, interacting with databases but also data frames
- Prepare, ingest and query large datasets
- Build cloud data pipelines

Extend DuckDB with custom functionality

DuckDB in Action introduces the DuckDB database and shows you how to use it to solve common data workflow problems. It’s full of quick wins—right from chapter one, you’ll be finding new ways that DuckDB can speed up your work as a data professional. Each new concept is paired with a hands-on project example, so you can easily see how DuckDB works in action.

about the book

DuckDB in Action will show you how to quickly get your hands dirty with DuckDB. You won’t need to read through pages of documentation—you’ll learn as you work. Begin with DuckDB’s CLI embedded mode, then dive straight into modern SQL queries and utilizing DuckDB’s handy SQL extensions. From there, you’ll explore the different ways you can analyze data with DuckDB, including advanced aggregation and analysis, data without persistence, and DuckDB’s underlying architecture. Learn how to combine DuckDB with the Python ecosystem for even greater customization, and how to extend DuckDB with its own tools. You’ll take to DuckDB like a duck to water, rapidly solving almost any relational data task with zero friction.

Genres: Software, Programming, Computer Science, Coding

312 pages, Paperback

Published September 1, 2024

Reviews

[Sebastian Gebski · Rating 3 out of 5]

Good tech, unimpressive book.

Yes, it does equip you with the basic knowledge (how to install, how to run the query, how to connect from your fav programming language, how to integrate with BI, how to import some data), but it fails (or rather skips or skims quickly) where you need expertise most:
- designing schemas for various scenarios
- in-depth analysis of query plans (there's a dedicated chapter on that, but frankly - it feels very rushed and I didn't learn much there)
- troubleshooting/optimization
- monitoring (observability)
- upscaling

In exchange, you get a chapter on Motherduck, which seems more like a bit of marketing than actual value ... 2.5 stars, rounded up, because it's the only book on DuckDB I know.

[Andrew Breza · Rating 4 out of 5]

A useful overview of DuckDB. Enough to get you started, and the book does a good job of highlighting what makes DuckDB different from other databases. The last several chapters include irrelevant things like introductions to Dagster and Superset. I wish those chapters had instead been dedicated to more in-depth features of DuckDB.

[Peter Aronson · Rating 3 out of 5]

This book is readable and covers a fair amount of ground, but ultimately it's kind of shallow to my tastes. It only gets into what I consider the meat of the matter in the last two chapters and the appendix. I like database books that start with the architecture before going into the use cases. It also didn't talk at all about writing extensions, which when dealing with an extendable DBMS is one of the most interesting things you could cover. It also spends too much time on specific tool sets that a particular user might or might not actual use. And the focus on Python is, I suspect, not the kind of thing that will age well.

It does do a nice job of covering DuckDB's rather idiosyncratic dialect if SQL though.

[Abhilash · Rating 5 out of 5]

It is a comprehensive guide to learn about DuckDB and has a lot practical examples. It is an invaluable resource for data professionals seeking to harness DuckDB's power

[F3 Sumo · Rating 5 out of 5]

Excellent dive into the capabilities of DuckDB. It's both highly informative and approachable, touching on a vast array of topics and use cases.