DROP DATABASE – a Galactic Parable
summary
A developer argues against using databases like SQL, JPA and Hibernate, claiming they severely hurt performance, code clarity, and data value. The talk demonstrates code for a system built without DBs, compares it to implementations using JPA/Hibernate (including benchmarks), and advocates keeping data in RAM instead of tables or JSON. Source code and comparisons are available at the provided GitHub repo.
AI-written summary · about
chapters (18)
- 0:00Introduction
- 2:22Multi-fork universe concept
- 4:10Galactic pizza model
- 7:21Java implementation
- 11:57Domain modeling in Java
- 15:53Java flaws
- 18:20SQL and Hibernate alternative
- 21:59Performance comparison
- 25:57Production deployment struggles
- 30:13Power outage recovery
- 33:45Business expansion challenges
- 35:33Technology history
- 37:17Cargo cult critique
- 39:02Simplified persistence
- 41:10Transaction basics
- 42:16CQRS and event sourcing
- 43:41Event-driven modeling
- 44:43Summary
related talks
49mDo You Really Need Hibernate?
35mAn Ode to 17 Databases in 39 minutes
51mEmpowering Enterprise Architectures with Polyglot Persistence
38mAn Ode to 17 Databases in 33 Minutes
43mCaching and Beyond - Smarter Data Processing with Java
34mPanel Discussion: Databases
46mThe State of Java Relational Persistence
44mAn ode to 17 databases in 29 minutes