Using Databases to pull your applications weight

on YouTube ↗31m

" summary

Shows concrete changes and measured wins: move heavy filtering and JSON assembly into Postgres, use materialized views for repeated reports, and convert queries to JSON so the database returns ready payloads. Using Postgres plus a small helper like the Surus gem reduces an API from ~2s to ~200ms and raises throughput from ~25 to ~467 requests/sec. Index judiciously, preload arrays to avoid many objects, and disable synchronous commits for noncritical writes to speed ingestion at the cost of possible recent data loss.

AI-written summary · about

" chapters (11)

1. 0:00Introduction
2. 3:06The case for database tools
3. 4:45Understanding query planners
4. 6:11Optimizing with EXPLAIN
5. 8:47Indexing strategies
6. 11:43Attribute preloading
7. 13:10Utilizing database views
8. 17:35Benchmarking and tradeoffs
9. 20:06Generating JSON in database
10. 25:46Synchronous commits
11. 28:52Conclusion and Q&A

" related talks

• 41mUsing Databases to Pull Your Application WeightHarisankar P S · Southeast Ruby 2017
• 46mUse PostgreSQL, not Rails, to make Rails Faster!Dan McClain · PostgresOpen 2014
• 42mHolistic Database Application TuningHenrietta Dombrovskaya, Grant Evans · PostgresOpen 2014
• 31mMaking Data DanceClark Barrett · RailsConf 2015
• 43mJust Use Postgres.Rhys Elsmore · PyCon Australia 2015
• 01h 01mNew Approach to Resolve objectrelational impedance mismatchHenrietta Dombrovaskaya · PostgresOpen 2013
• 30mPuny to Powerful PostgreSQL Rails AppsAndrew Atkinson · RailsConf 2022
• 23mFANTASTIC DATABASES AND WHERE TO FIND THEMChris Hasinski · Euruko 2023