Docker is a platform on which developers can build and deploy software using containers. It has gained a lot of popularity over the last few years. Docker provides methods that help a developer to build software that is productive; developers can also benefit from Docker's bit-for-bit compatibility between different environments. Docker has now become an important tool for solving the constant flow of deliveries that are common within modern infrastructure. Note that the containers used need to have a new, more refined monitoring approach, so if a developer uses Docker, they should opt for the newest integration available—Datadog.

Although this integration helps you monitor containers, it is easier to run version 4.3.1 of the Datadog agent.

If a developer requires a straightforward way to monitor Docker containers, they should be running a Datadog agent on the host. From here, the agent gains access to the statistics of containers, which allows Docker to be deployed on operating systems of recently running hosts, with current applications such as databases.

Docker uses current kernel constructs such as cgroups and namespaces to run containers. This is why the Datadog agent employs the accounting metrics of the local cgroup to collect memory, I/O, network, and CPU metrics that are relevant to the containers. The process is done every 15 seconds like clockwork before the information is sent out to Datadog. This method is the simplest, but there is also a Dockerized version of the agent that is available for developers who want to use containers to run an entire piece of software.

In Datadog, developers define metrics using dashboards and graphs, which are based on tags. This technique allows a developer to track down the relative metrics for a lot of containers, and tags can then be utilized to create a graph.