Open
Description
We could provide statistics views that plot various aspects of the system.
- Show the average/nth percentile latencies of services/RPC endpoints
- Show the error rate
Two kinds of view come to mind: A view for the entire system with tables, and a per-trace view with graphs and some configurable window size.