Workshop: Analyzing and Visualizing OpenTelemetry Traces with SQL
Talkat OTel Community Day North America 2022 · on Monday, 2022-06-20· at 13:50· 303+304, JW Marriott Austin | 110 E 2nd St·
summary
Hands-on workshop shows traces ingested via the OpenTelemetry Collector into Promscale / Timescale Postgres (docker demo) and explored with Grafana + Jaeger using plain SQL. Queries use time_bucket & percentile_ag to produce throughput, histogram, and p95 over time (example: ~712 traces in 5 minutes, peak 46 traces/10s, slowest trace 7.2s). A recursive CTE for tree walk builds upstream/downstream topologies and a node graph that pinpoints the bottleneck (digit service random digit: 10.4 minutes total execution, p95 ~4.2s). All examples run live from the demo repo.
AI-written summary · about
chapters (18)
- 0:00Start
- 0:36Analyzing and Visualizing Open Telemetry Traces with Sql
- 2:17Demo System
- 4:49Open Telemetry Collector
- 8:21Attach Events to a Span
- 11:49Build a Dashboard
- 17:35Time Buckets
- 20:31Histogram of Latencies
- 22:04P95 Latencies
- 24:14Build a Histogram of Latencies over Time
- 28:31Time Filter
- 28:56Edit Enable the Query
- 36:20Sampling
- 39:10How To Count to 10 and Sql
- 40:07Recursion
- 40:47Sequel Can Be Recursive
- 49:12The Node Graph
- 1:06:40Google Forms Survey
related talks
35mOpenTelemetry for Busy People
02h 03mWorkshop 3
41mModern Application Debugging: An Introduction To OpenTelemetry
37mOpenTelemetry & Distributed Tracing: A Jaeger v2 Deep Dive
13mConnecting Prometheus and OpenTelemetry Data for Faster Troublesho...
38mPractical introduction to OpenTelemetry tracing
32mDistributed Tracing in Action: Our Journey with OpenTelemetry
09mGraduated Project Lightning Talk: Using Trace Data for Monitoring & Alerting of Appl...