How do I collect CPU metrics per core?

By default the Datadog agent collects aggregate CPU stats, however, it's possible to enable per-core metrics by simply renaming a single conf file.

  1. Locate your conf files directory
  2. Locate and rename `system_core.yaml.example` to `system_core.yaml`
  3. Restart your agent

To begin using the new core metrics:

  1. In this example I'll create a new TimeBoard
  2. Under "Get", if you begin to type `system core` you'll immediately see a list of metrics that you may use, in this example I've selected `system.core.system`
  3. Optionally, you may now select things like a host, a set of instances, specific cores using the "From" dropdown, I've selected an individual host for this example
  4. In the last dropdown you can select "core"

For additional core specific metrics and details, reference: http://docs.datadoghq.com/integrations/systemcore/

 

 

 

Have more questions? Submit a request

0 Comments

Please sign in to leave a comment.
Powered by Zendesk