Building a Performance Analytics Tool with ActiveSupport

on YouTube ↗21m

" summary

Builds a tiny production-ready performance analytics pipeline called active monitoring that actually records response times and each SQL query using ActiveSupport instrumentation, writes metrics to InfluxDB time-series storage and graphs them on a dashboard (Grafana). Uses ~50 Rails hooks, implements a Rails plugin, a current_attributes request cache to join SQL events to requests, normalizes SQL with regexes before grouping, and relies on LSM-tree retention/downsampling. Notes measurable overhead from normalization and suggests moving it to C or async jobs if latency becomes significant.

AI-written summary · about

" chapters (7)

1. 0:00Introduction
2. 0:40Overview
3. 2:56ActiveSupport
4. 5:38Metrics
5. 12:18Processing
6. 16:02Visualization
7. 19:45Summary

" related talks

• 26mRails Performance Monitoring 101: A Primer for DevelopersRishi Jain · RubyConf TH 2023
• 30mRails Performance monitoring: A primer for developersRishi Jain · RubyConf India 2023
• 45mRails Performance Monitoring 101: A Primer for Junior DevelopersRishi Jain · RailsConf 2023
• 36mIt's not in Production Unless it's MonitoredJoseph Ruscio · RailsConf 2012
• 01h 02mPerformanceJeremy Kemper · RailsConf Europe 2008
• 17mActiveSupport::Notifications and Live Status PagesMatthew Delves · Red Dot Ruby Conference 2014
• 54mOptimizing performance in Rails apps with GraphQL layerCaio Almeida · wroc_love.rb 2024
• 30mBuilding RailsPerf, a toolkit to detect performance regressionsKir Shatrov · RailsConf 2015