APM and Services

apm and services is a dedicated area for quickly analyzing the performance of applications and services in terms of requests, errors, latency, and throughput services overview page the overview page lists all instrumented applications/services in a tabular view, allowing you to compare their health at a glance each row represents a single service and includes service name of the application or service requests total number of requests handled by the service for the selected time range throughput rate of incoming requests (for example, requests per minute), indicating overall load on the service p99, p95, p50 latency tail and median latency percentiles that show how long requests take to complete p99 latency at the 99th percentile (worst‑case performance for most users) p95 latency at the 95th percentile p50 median latency, representing a typical request error rate percentage of failed requests, helping you quickly identify unstable services version reported version of the deployed application or service, useful for correlating performance with releases apm dashboard for a service clicking any service in the overview opens its apm dashboard , a preconfigured view focused on that specific service the dashboard shows requests, errors, and latency over time timeline charts that help identify spikes in traffic, error bursts, or latency regressions destinations table a breakdown of downstream dependencies that the service communicates with, including destination name requests per second (req/s) error percentage for calls to that destination latency metrics between the service and each destination related issues, alarms, and dashboards panels that surface existing issues and alarms for the service, and quick links to any additional dashboards created for deeper analysis traces and logs for deep analysis from the apm dashboard, you can inspect related traces and logs for that service traces table shows recent traces with fields like start time, span name, kind (client/server), duration, and an action to view the full trace for end‑to‑end request analysis logs table lists recent logs with time, severity, and message, along with actions to explore logs in detail and correlate them with traces and issues for faster root‑cause investigation