Proposal

Right now, the Datadog scaler polls the Datadog REST API to get the query values that will drive scaling events. This solutions works, but many users, particularly those who have big infrastructures, are getting rate limited, limiting the usefulness of the feature.

The Datadog Cluster Agent, a component that is present in clusters that are monitored with Datadog, already has functionality to get metric values from Datadog with the same purpose (autoscaling).

Ideally, the Datadog KEDA scaler would have an option to connect to the Cluster Agent to get metrics values, instead of calling the REST API, as it will decrease the risk of being rate limited.

The user would have the choice of choosing either the REST API (that would remain default for now, for backwards compatibility), or choose to connect to the Cluster Agent instead.

Use-Case

No response

Is this a feature you are interested in implementing yourself?

Yes

Anything else?

No response