Add Compute and Storage Workers

Layer the compute-worker values on top of the controller values and run the Helm upgrade. Compute workers require the host access needed to execute workloads, so their node selection and device/network requirements are not identical to the controller.

Add the storage values layer when the selected node will contribute storage. If the intended topology is controller plus storage only, omit the compute worker layer instead of enabling a compute role you do not intend to use.

Running controller and worker roles on the same node can be useful for a lab but should be an explicit choice with independent role labels. Keep production placement policy separate from a compact demonstration topology.

Replace example node, registry, namespace, and endpoint values with the values for the target environment.

The enabled compute and/or storage worker pods appear on the intended nodes and enroll to the running controller.