# Server Metrics

Server metric messages provide telemetry data from the NATS server for monitoring and performance analysis.

## Overview

Server metrics are published as events that can be subscribed to for real-time monitoring. These metrics help track:

* Service response times
* Request processing latency
* System performance indicators
* Resource utilization

## Available Metrics

### [Service Latency](/reference/system/metric/service-latency.md)

Measures the time taken to process service requests:

* Request/response round-trip times
* Processing latency distribution
* Performance bottleneck identification

## Subscribing to Metrics

Metrics are published to specific subjects that follow the pattern:

* `$SYS.SERVER.METRIC.>` - All server metrics
* `$SYS.SERVER.METRIC.SERVICE.LATENCY` - Service latency metrics

## Best Practices

* Use sampling to reduce metric overhead
* Store metrics in time-series databases
* Set up alerts for latency thresholds
* Correlate metrics with system events