Skip to main content
Performance Monitoring in Production

Performance Monitoring in Production

Andrius LukminasAndrius LukminasDecember 16, 20254 min read395 views

What We Monitor

Key performance indicators we track:

  • Page Load Time - Time to interactive
  • API Response Time - P50, P95, P99 latencies
  • Database Query Time - Slow query detection
  • Error Rate - Errors per minute
  • Memory Usage - Node.js heap monitoring

Our Stack

  • Custom Metrics - Tracked in error_logs table
  • Console Monitoring - Structured logging
  • Admin Dashboard - Real-time metrics display

Performance Budget

We set targets and alert when exceeded:

  • Dashboard load: < 1s
  • API endpoints: < 200ms P95
  • Database queries: < 100ms

Optimization Workflow

  1. Monitor metrics continuously
  2. Alert on threshold breaches
  3. Profile to identify bottlenecks
  4. Implement fixes
  5. Verify improvement

Recent Wins

By adding database indexes and optimizing Prisma queries, we reduced average API response time from 450ms to 120ms.

Related Articles

Comments

0/5000 characters

Comments from guests require moderation.