A Practical Guide to Dashboarding
summary
Walks through a working pipeline built with statsd and Graphite, Docker (Katacoda), and Grafana visualization that ingests a tiny open‑source response API returning HTTP codes so metrics show count, mean duration and error by status. Applies RED and Google golden signals plus a USC-style service checklist. Shows a real incident: 78 errors traced to Kinesis iterator saturation (42 minutes) and fixed in about 10 minutes. Describes ECS production setup using Whisper, EFS, Postgres and CloudWatch caveats.
AI-written summary · about
chapters (17)
- 0:00Start
- 0:18What Is a Dashboard and Why
- 0:45Favorite Dashboard
- 1:26Speedometer
- 5:11Utilization
- 5:37Saturation
- 9:50Latency
- 11:25Server Service Level Agreements
- 12:00Rate Error and Duration
- 13:16Response Api
- 17:33Create a Network
- 19:53Data Source
- 24:44Visualization
- 27:04Alias
- 36:50Errors of 78
- 40:06Resources
- 41:19Practical Monitoring by Mike Julian
related talks
49mA Practical Guide to Dashboarding
28mProduct metrics for developers
45mCreating & Maintaining Impactful Dashboards'
11mBuilding lightweight observability system for Lean Teams
21mJourney from Broken Dashboards to O11y Maturity
38mThe Structure and Interpretation of Graphs
38mA Practical Guide to Prometheus for App Developers [B]
25mFrom Scratch to Production-Ready: Creating Dashboard Visualizations