Creating an inhouse library to measure performance
summary
Built an in-house performance library on top of Firebase to measure TTP and TTI plus arbitrary custom spans, exposing a synchronous-looking JS API that actually batches work and makes only two native bridge calls per trace (start and stop). It records span names and durations, sends JSON traces to existing analytics, and surfaces regressions. Outcome: meaningful device- and network-aware metrics, early detection of regressions during a full UI revamp, and fewer user slow-app complaints.
AI-written summary · about
chapters (6)
- 0:00Introduction and motivation
- 2:37Key metrics for measurement
- 4:46Existing solutions and decision
- 5:31Implementation details
- 9:17Behind the scenes design
- 11:04Impact on organization
related talks
30mImprove app performance and stability with Firebase
18mThe Facebook Mobile Performance Tooling Ecosystem: An Overview
16mCoding Session: Stunning Performance
55mBuilding Quality with Firebase
28mMeasurement and Analysis for Sustainable Service Delivery
16mFrom Beginner to Expert in Web Performance
44mPractical guide to performance Benchmarks
20mReal metrics for front-end performance