Germany's Next Top Data Model

on YouTube ↗18m

" summary

Argues that traditional normalized schemas fail at scale and NoSQL forces query-driven design, so align storage to application queries rather than forcing joins. Recommends using denormalization and timeboxing (store an hour or 15-minute bucket as one document) to cut cross-node queries, prefer natural keys over auto-increments to avoid extra lookups, and use convergent/commutative replicated data types (CRDTs) and tiered stores (raw durable + in-memory presentation) to keep models scalable and fast.

AI-written summary · about

" chapters (6)

1. 0:00Intro
2. 1:40The problem with relational databases
3. 3:40How to model data
4. 6:55Denormalization
5. 11:45Scaling
6. 14:20Time Boxing

" related talks

• 26mNoSQL Data Modeling for Front-end DevelopersStanimira Vlaeva · EnterpriseNG 2021
• 34mData Storage: NoSQL Toasters and a Cloud of Kitchen SinksCasey Rosenthal · RailsConf 2013
• 38mNot all data is created equal: NoSQLSantosh Hari · DevFest Florida 2019
• 01h 00mFrom Relational Databases to NoSQLScott Davis · Great Indian Developer Summit 2015
• 42mSay YES! to NoSQL: A Guide on When to Ditch Relational DatabasesAdrienne Braganza Tacke · AzConf 2020
• 41mReal-World NoSQL Schema Design #bbuzzTugdual Grall · Berlin Buzzwords 2016
• 20mDomain Driven Design & NoSQLLucas Dohmen · Red Dot Ruby Conference 2014
• 31mSQL to NoSQL to NewSQL and the rise of polyglot persistence Paul DixPaul Dix · GORUCO 2017