In case your system requires different nodes or processors to execute, you should explain the mapping of your building blocks to that hardware.

Document or specify the mapping of the building blocks (see arc42 section 5) onto the hardware (more specifically - the mapping of the artifacts generated/compiled/created from the actual source code building blocks.). That is called deployment.
In many cases that can be an m:n mapping, with several variants of deployment artifacts

Remember to explain these deployment variants.

In the following diagram you find three different variants how the three architecture building blocks A, B and C can be deployed on different hardware environments.

three different variants how the three architecture building blocks A, B and C can be deployed