zoom_in Click to enlarge

Datadog Agent

▲ ▼

description Datadog Agent Overview

While Datadog is the platform, the Agent itself is a critical component. This lightweight, unified agent is responsible for collecting metrics, logs, and traces from the host OS, containers, and applications. Its strength is its ability to collect diverse data types using a single, standardized mechanism, reducing operational overhead significantly for the user.

help Datadog Agent FAQ

What does the Datadog Agent collect from a server?

The Datadog Agent collects host metrics, process data, logs, traces, and integration data depending on how it is configured. It is the component you install on Linux, Windows, Kubernetes nodes, or cloud instances before the data appears in Datadog dashboards.

Why do teams run one Datadog Agent instead of separate log and metrics shippers?

Datadog's Agent is unified, so one installed service can handle infrastructure metrics, log forwarding, APM trace intake, and checks such as NGINX, PostgreSQL, or Redis. That reduces the number of local collectors compared with mixing tools like Telegraf, Fluent Bit, and a separate tracing agent.

What changed when Datadog moved from Agent 5 to Agent 6 and 7?

Datadog Agent 6 was a major rewrite in Go after the older Python-based Agent 5 generation. Agent 7 continued the Go-based line and removed the embedded Python 2 runtime, which mattered for custom checks and older integrations.

Does the Datadog Agent work in Kubernetes?

Yes, Datadog commonly runs the Agent as a DaemonSet in Kubernetes so each node has a collector. It can also use the Datadog Cluster Agent for cluster-level metadata, autoscaling support, and reducing direct API-server load.

swap_horiz

Looking for Datadog Agent alternatives? Compare top competitors ranked & scored

arrow_forward