Demystifying Database Performance Issues with SQL Commenter
summary
An open source library called SQL Commenter injects URL‑encoded, lexicographically sorted key=value comments into SQL so logs and APMs can link queries back to application context; it auto‑instruments about eight to ten ORMs across four languages (Python, Java, Node.js, Ruby) and targets Postgres/MySQL/MariaDB/SQLite and Cloud SQL. Cloud SQL Insights visualizes tags and sampled query plans; trace context can be sampled (adjust sampler to increase captures). Donated to OpenTelemetry recently; overhead is reported as small (~12 MB for default settings). action=assign_order traceparent included
AI-written summary · about
chapters (5)
- 0:00Start
- 0:57Jan Kleinert
- 6:09What Do We Support Today
- 16:41Cloud Sql Insights
- 29:08Sampling Probability Sampler
related talks
26mDemystifying Database Performance Issues with sqlcommenter
18mThe Future of Database Observability is Open: SQL Commenter Merges W...
22mCorrelating Application and Database Performance Using OpenTelemetry
46mAwesome SQL – tips and tricks for application developers
18mApparently, you can "debug" your SQL queries
52mThe SQL inside the ORM/Builder box | Claudio Zizza
34mThe State of SQL-Based Observability
15mOLTP SQL Database Query Tracing and Linting