docs/sources/datasources/parca/_index.md
Parca is a continuous profiling database for analysis of CPU and memory usage, down to the line number and throughout time. Grafana ships with built-in support for Parca, so you can add it as a data source and start querying your profiles in Explore.
Refer to the Parca documentation to learn about continuous profiling and how to instrument your applications.
To use Parca profiling data in Grafana, you should:
This data source supports Parca v0.19 and later.
| Feature | Supported |
|---|---|
| Profiles | Yes |
| Metrics | Yes |
| Alerting | No |
| Annotations | No |
The following pages help you get started with the Parca data source:
Explore is the primary way to interact with Parca data in Grafana. Use Explore to query profiles and metrics without building a dashboard, and to select the Both query type to view flame graphs and time-series data side by side.
Using the Parca data source, you can embed profiling data in your dashboards alongside other signals like logs and metrics. For example, you can place a flame graph panel showing CPU profiles next to a metrics panel tracking request latency to correlate performance bottlenecks with application behavior.
Always ensure that your plugin version is up-to-date so you have access to all current features and improvements. Navigate to Plugins and data > Plugins to check for updates. Grafana recommends upgrading to the latest Grafana version, and this applies to plugins as well.
{{< admonition type="note" >}} Plugins are automatically updated in Grafana Cloud. {{< /admonition >}}