Worker Node

A Worker Node is compute dFlow uses to run your Services. Worker Nodes are not a replacement for Applications, Environments, or Services; they are the machines (or managed instances) attached to an Environment so workloads have somewhere to build and run.

How Worker Nodes fit the model

• The control plane (dFlow dashboard and API) plans what should run.
• Worker Nodes perform builds, host containers, and expose logs and metrics for the Services in the Environments they serve.

You attach one or more Worker Nodes to an Environment (or use managed compute on dFlow Cloud where your plan supports it). All Services in that environment that need compute share that attachment context.

Connection model

For servers you operate, dFlow typically connects over SSH and prepares the runtime (commonly Dokku for container-style app deployments). You do not need a permanent agent on every machine for that path. See Add a self-hosted server under Worker Nodes and Compute in the sidebar for the onboarding task.

See also

• Platform architecture: control plane, proxy, builds, and monitoring.
• Environments: where compute attaches.
• How dFlow is structured: full hierarchy.
• Worker Nodes overview: section depth when published.
• SSH keys: credentials for node access.