// service map Live dependencies
Healthy Error
checkout-api → inventory-api · 2 errors // real product UI
Not a mockup - the running app.
The animation above shows the idea. This is the actual workspace view.
// glance, then drill
Read dependencies and health at a glance.
A user reports the checkout is slow. Where do you look first? Open the map: the edge from api to payments is red at p95 312ms, while everything else is green. You have not opened a single trace yet - and you already know which call to drill into.
- Trace-derived service dependency graph
- Edge latency summaries
- Edge error summaries
- Edge request and count summaries
- Filter by service, environment, and host
// arrange once, persist forever
Lay it out once, drill in any time.
Drag your services into the shape your team thinks about them - frontend at the top, queues on the side, databases at the bottom. The layout stays put across sessions. Click any edge to jump straight into a representative trace.
- Pan, zoom, drag, and reset controls
- Persisted node positions
- Sample trace drilldown from any edge
// built on
What feeds it.
// next step
Watch your map build itself.
Send traces and the dependency graph appears automatically.