The Weirdest PromQL You’ll Ever See: PromQL for Reporting, Analytics
summary
Uses Prometheus + Grafana to do billing and usage analytics and reports $25,000/month of metrics dialed back in testing. Shows concrete queries: use increase(counter, 1d) for counters, integrate rate metrics by using sum_over_time(range) scrape_interval inside subqueries to avoid missing scrapes, and isolate calendar months by applying a month() filter via absent() and group_left ignoring* join so labels are preserved. Demo works and supports unioned months for multi-month views.
AI-written summary · about
chapters (11)
- 0:00Introduction to cost analysis
- 1:14Why use Prometheus for billing
- 3:53Defining cost visualization goals
- 7:10Building daily usage reports
- 9:28Handling gauge metrics
- 12:27Filtering for monthly data
- 16:04Combining metrics with joins
- 18:05Creating cumulative views
- 20:26Aggregating multiple months
- 22:58Demonstration and closing
- 24:09Q&A session
related talks
09mFun and Games: Using Prometheus to aggregate disparate metric dat...
28mGrafana is Not Enough: DIY User Interfaces for Prometheus
12mLightning Talk: A Drop-in System To Accelerate Metrics Observability by 100x Us...
31mMaking Data Dance
22mBuilding Interactive Grafana Dashboards for Business Intelligence
33mMaking Data Dance
46mTaking the Scare out of Monster Reports
07mPortfolio Marketing Metrics and Insights