Breakout Session
summary
Heroku Postgres runs about a million databases on AWS and uses LXC containers to keep runtime images uniform while storing data separately, grows disks via EBS elasticity, archives WAL segments with Wally to S3 for point‑in‑time rebuilds, and builds hot‑standbys and forks from those backups. Monitoring and automated incident resolution restart resources or fail over followers, and the platform aims for immutable-ish servers so upgrades and large‑scale recovery work reliably.
AI-written summary · about
chapters (11)
- 0:00Start
- 1:20What is Heroku?
- 5:05Applications
- 6:19heroku-postgresql heroku-redis
- 14:30Automated incident resolution
- 15:47Stuff happens constantly
- 16:29Circuit Breakers
- 20:12Immutable-ish Infrastructure
- 22:05Durability and Availability
- 28:41complicated project
- 29:33composable services
related talks
22mLessons from Automatic Incident Resolution for a Million Databases
46mBig databases, many servers, data center or cloud - it's all here!
35mContainerizing PostgreSQL and Making it Cloud Native Ready
36mAutomated Performance Testing of Postgres
19mPostgreSQL at Amazon
51mIdentity at Scale; How Okta uses Postgres
43mJust Use Postgres.
11mGiving your Heroku App highly-available PostgreSQL
