Reading DDIA

Why this book

The book data engineers keep recommending

DDIA explains how data systems actually work underneath: storage engines, replication, partitioning, transactions, and batch and stream processing. It doesn't teach any single tool. It teaches the ideas every tool is built from, which makes it easier to judge a new database than reading its marketing page.

I'm reading the second edition, updated for cloud-native systems, and writing up each chapter as I go. Writing forces me to check that I actually understood it.

How each post works

Summary, take, push back

• The gist: what the chapter argues, in plain language.
• My take: what I found most useful, tied to real work where it fits.
• Push back: where the chapter is thin, oversimplified or in tension with itself.

Chapters

Notes so far