Database Wizardry for Legacy Applications
summary
A MySQL report query that stalls for eight seconds becomes 35 milliseconds after creating a MongoDB replica for reads and keeping recent records there; a text-search on a materialized path drops from 16 seconds to 0.3 seconds by applying an added level filter instead of LIKE; heavy inserts are asynced to a queue or cron job for eventual synchronization; large free‑text blobs move out of the relational store into S3 + CloudSearch to cut DB load and storage costs.
AI-written summary · about
chapters (17)
- 0:00Introduction
- 0:53Who am I
- 2:36Respect
- 3:59Legacy Applications
- 5:31Database Optimization
- 6:00Database Scaling
- 6:45Database Structures
- 8:28Binary Tree
- 12:49Improve Database Performance
- 13:50What to Learn
- 15:18Money Transactions
- 15:53Dimensional Tables
- 19:11QueryMongo
- 25:02Passive Aversion
- 27:08The Best Solution
- 34:58Text Search
- 36:04Conclusion
related talks
41mUsing Databases to Pull Your Application Weight
01h 50mSQL Execution Internals – From Database to OS and Hardware Levels
58mWhy Databases Cry At Night
35mDatabase Stalls, From the Ordinary to the Obscure
27mMurdering our Mongo Database (or how we migrated to a relational database)
37mNavigating Managed Databases for Over a Decade as a Non-DBA
30mOptimising your Database for Analytics
44mMongoDB to MySQL the How and the Why