Before you can use AMD GPUs in {productname-short}, you must install the required dependencies, deploy the AMD GPU Operator, and configure the environment.
-
Install the latest version of the AMD GPU Operator, as described in Install AMD GPU Operator on OpenShift.
-
After installing the AMD GPU Operator, configure the AMD drivers required by the Operator as described in the documentation: Configure AMD drivers for the GPU Operator.
Note
|
Alternatively, you can install the AMD GPU Operator from the {org-name} Catalog. For more information, see Install AMD GPU Operator from Red Hat Catalog. |
-
After installing the AMD GPU Operator, create an accelerator profile, as described in Working with accelerator profiles.
From the Administrator perspective, go to the Operators → Installed Operators page. Confirm that the following Operators appear:
-
AMD GPU Operator
-
Node Feature Discovery (NFD)
-
Kernel Module Management (KMM)
Note
|
Ensure that you follow all the steps for proper driver installation and configuration. Incorrect installation or configuration may prevent the AMD GPUs from being recognized or functioning properly. |