Monitoring tools should be flexible, so we provide multiple ways to interact with Datadog. This section gives you a quick overview of the various methods available
Datadog features a powerful API to programmatically access the service. You can submit custom metrics; manage dashboards and monitors; schedule downtimes; and perform administrative actions. Available endpoints can be found on our dedicated API documentation page.
In addition, many libraries are available to directly interact with the API.
While you can submit metrics directly through our API, you can also submit metrics via the Datadog Agent using DogStatsD and custom Agent checks.
The Datadog Agent includes DogStatsD, a powerful statsd daemon with additional features, that provides better control over your metric metadata and aggregation. Multiple libraries are available to easily send metrics from your application to Datadog using DogStatsD.
The API and DogStatsD are handy to push metrics to Datadog from one of your application. If you would like to regularly pull metrics from a particular setup, similar to what is done by the Agent for the systems we directly support, you can add your own integration. We have provided detailed steps and resources for developing and testing new integrations. These integrations can then be easily shared with the Datadog community, for instance through our Datadog/integrations-extras repository
When sending metrics via DogStatsD or a custom integration, it is helpful to have a deeper understanding of metrics. Here are some technical resources regarding metric internals:
You can also find our APM tracing utilities, which you can use and contribute to. Support on new languages can be proposed by the community, along with community APM integrations.
Many users from our community have already shared integrations with a large number of commonly used applications. These resources are also useful to reference when starting the work on a new Agent integration.
