Not on Our Watch: an Introduction to Application Performance Monitoring
summary
Enterprise APM New Relic is used to correlate alerts, deployments and transaction traces and reveals a root cause: a Drupal 7 filesystem with directories holding tens of thousands of files (a Drupal 7 file bucket issue) that triggers persistent warnings despite sub‑second averages. Fixes to queries and file distribution cut median page time from ~700ms to ~200ms — a 71.4% decrease — and the talk stresses apdex thresholds (~0.7–0.8), deployment tagging and choosing APM by language, cost and custom metric support.
AI-written summary · about
chapters (9)
- 0:00Welcome and introduction
- 1:07Performance expectations over time
- 8:01What is APM?
- 11:15Case study solving performance issues
- 16:41Key application performance metrics
- 28:06Components of APM solution
- 31:23Best practices for APM
- 38:31Do you really need APM?
- 41:50Q&A and real-world fixes
related talks
30mRails Performance monitoring: A primer for developers
38mApplication Performances Monitoring with Elastic APM and the ELK Stack
24mFull Observability for Application Monitoring
38mEffective App Monitoring in Production
26mRails Performance Monitoring 101: A Primer for Developers
47mUsing APM to Give Businesses a Competitive Edge
32mDevOps and Agile: AppDynamics in Continuous Integration Environments
41mMeasuring the Moving Parts of User Experience in Java/.NET applications