Free HP HPE7-S02 Practice Exam with Questions & Answers | Set: 2

D is correct. A NIM image hosted in NVIDIA NGC is protected by NVIDIA registry authentication. To allow KServe in a user namespace to pull that image, the user needs a valid NGC API key and a Kubernetes secret containing the corresponding registry credentials in the namespace where the InferenceService will run. The workload can then reference that secret for the image pull or other required NGC access.

Option A describes registering NGC with HPE Machine Learning Inference Software. That is relevant when MLIS is the serving mechanism, but the scenario explicitly uses KServe. Option B is not how NVIDIA AI Enterprise is made available to HPE AI Essentials; it is not imported as a user Helm chart. Option C is also invalid because KServe does not require special “decorators” in application code to launch a NIM.

Namespace placement matters because Kubernetes secrets are namespace-scoped. A secret created elsewhere is not automatically available to the user’s KServe deployment.

References/topics: Advanced HPE Compute Solutions, Rev. 26.21, Module 5 “Other options for serving models on HPE Private Cloud AI” and NVIDIA AI Enterprise; HPE AI Essentials KServe guidance; NVIDIA NGC authentication documentation.

References/topics: Advanced HPE Compute Solutions, Rev. 26.21, Module 5 “Serving NIMs with HPE Machine Learning Inference Software”; HPE AI Essentials/MLIS registry documentation and release-history notes.