Prepping Your Project for Production
summary
Recommends starting on a PaaS (Heroku) or managed Kubernetes, and use managed backing services because a database outage can be game over; deploy with Gunicorn (5s timeout, configure workers) or uWSGI if needed; serve static with WhiteNoise and push media to S3 via django-storages with signed URLs for private files; set DB connection pooling like CONN_MAX_AGE, reduce queries with select_related/prefetch, and apply template fragment caching (example: cut a 2s nav render to ~200ms). Monitor with New Relic/Scout/Datadog and error-report with Sentry; lock dependencies with pip-compile.
AI-written summary · about
chapters (9)
- 0:00Introduction
- 1:45Choosing a hosting platform
- 7:49Backing services
- 9:22Application configuration
- 15:09Web servers
- 18:10Managing assets and media
- 23:51Performance optimization
- 31:40Security best practices
- 38:53Observability and conclusion
related talks
21mPractical Deployment Strategies for Production Python Apps
26mProtoype for Production
26mWhat Django Deployment is Really About
30m"What Django Deployment is Really About"
41mProduction Django: Building a Highly Scalable, Secure Django Site
21mHigh-Availability Django
44mHow to Write Deployment-friendly Applications
32mChoose Your Own Django Deployment Adventure