Cache is King: Get the Most Bang for Your Buck From Ruby
summary
Shows concrete fixes and measurable wins: using Ruby to implement bulk serialization for 200M vulnerability records drops 300-item serializations from ~2100 MySQL calls to 7 and cuts time from ~6s to <1s, moving RDS CPU from ~100% to ~25%. Replacing repeated Redis gets with a local Ruby hash cache yields ~65% job speedup. Uses ActiveModel serializers, Elasticsearch, MySQL, Redis, octopus sharding gem, Datadog and Redis MONITOR. Eliminating needless datastore hits (skip empty reports — 25k reports, 10k blank — cuts nightly work from 10h to 3h).
AI-written summary · about
chapters (7)
- 0:00Start
- 0:36redis
- 1:39Demo Time!
- 4:23KENNA Security
- 7:53Bulk Serialization
- 16:34285 Workers
- 32:10Questions
