PostgreSQL Beyond Django: Strategies to Get Max Performance with Álvaro Justen
summary
Concrete claims and concrete numbers: a 52M+ row Django demo shows per-row inserts taking days, bulk_create ~2.5 hours and COPY to table via psql 8m20s (≈123k rows/s). Upgrading Postgres cuts table/index size ~5%. Changing ID strategy and downscaling integer types plus reordering columns reduces space (~3% each) and speeds access. Full‑text options compare precomputed search_vector, trigger maintenance, and an expression GIN index; search_vector + GIN index ≈150ms versus ~10s without. Also mentions building a simple Django app and using triggers and specialized indexes.
AI-written summary · about
chapters (14)
- 0:00PostgreSQL performance intro
- 1:03Real-world PostgreSQL use cases
- 3:10General PostgreSQL advice
- 7:56Why upgrade PostgreSQL
- 10:48Data modeling impact
- 14:23Optimizing primary keys
- 18:42Small integers for space
- 23:11Column order for storage
- 26:30Automate schema optimization
- 28:19Full-text search in PostgreSQL
- 33:24Advanced full-text indexing
- 37:29PostgreSQL features overview
- 39:58Efficient data ingestion
- 42:51Q&A
related talks
22mDjango with PostgreSQL superpowers
23mDjango with PostgreSQL superpowers
43mOptimize Python & Django apps with Postgres superpowers
21mDjango and PostgreSQL: An Even-Closer Union
29mMigrating Data Quickly and Safely
44mPostgreSQL in Django 1.8
42mPostgreSQL Tips & Tricks For App Devs
26mPostgreSQL Generated Columns by Example