NCP-AIO Workload Management Practice Question
An administrator runs a Kubernetes cluster with the NVIDIA GPU Operator. A data science team wants to run several small inference containers that each use only a fraction of a GPU's compute and memory, but the cluster currently assigns whole GPUs per pod. Which approach allows multiple containers to share a single physical GPU with memory isolation?
⚠ Common exam trap
The trap here is conflating time-slicing or MPS, which share a GPU without memory isolation, with MIG, which provides true hardware-level memory partitioning.
Answer choices
Why each option matters
Answer the question above first, then reveal the full breakdown to understand why each option is right or wrong.
Correct answer & explanation
✓
Enable Multi-Instance GPU mode on supported GPUs and configure the device plugin to advertise MIG instances as schedulable resources.
MIG is the only mechanism here that divides a physical GPU into hardware-isolated instances with separate memory and compute slices. By enabling MIG and having the device plugin advertise each instance, the scheduler can place multiple inference containers on one GPU while keeping their memory and faults isolated.
Answer analysis
Option-by-option breakdown
For each option: why learners choose it and why it is or isn't the right answer here.
- ✓
Enable Multi-Instance GPU mode on supported GPUs and configure the device plugin to advertise MIG instances as schedulable resources.
Why this is correct
MIG partitions a supported GPU into hardware-isolated instances with dedicated compute and memory slices. Advertising those instances through the device plugin lets each inference container receive its own MIG device, providing true memory and fault isolation while allowing multiple containers to share one physical GPU, which matches the requirement.
- ✗
Reduce the container's nvidia.com/gpu request to a fractional value such as 0.5.
Why it's wrong here
Kubernetes extended resources are integer-only, so a fractional nvidia.com/gpu request is rejected by the API server. Even if it were accepted, the device plugin would still allocate a whole device, so this approach neither works syntactically nor delivers the desired fractional, isolated sharing.
- ✗
Set the device plugin's time-slicing configuration so that each GPU is advertised multiple times.
Why it's wrong here
Time-slicing lets multiple containers share a GPU by interleaving execution, but it provides no memory isolation or fault isolation between them. A memory-exhausting or crashing container can still affect its neighbors, so time-slicing does not satisfy the explicit requirement for isolated memory per shared container.
- ✗
Configure the MPS control daemon to allow concurrent kernel execution across containers.
Why it's wrong here
MPS improves concurrency and utilization by allowing kernels from multiple processes to run simultaneously, but it does not partition memory into isolated slices. Containers still share the same memory space and can interfere, so MPS alone does not deliver the memory isolation the team requires for safe co-tenancy.
About these practice questions
This NCP-AIO question is part of Courseiva's 309-question bank — original exam-style content with full explanations and wrong-answer analysis, never real exam questions or exam dumps. Learn why practice questions differ from exam dumps →
JA
Written and reviewed by Johnson Ajibi, MSc IT Security
Senior Network & Security Engineer · founder of Courseiva
Last reviewed September 2026 · checked against the official NVIDIA exam blueprint
This NCP-AIO practice question is part of Courseiva's free NVIDIA certification practice question bank. Courseiva provides original exam-style practice questions with explanations, topic-based practice, mock exams, readiness tracking, and study analytics to help learners prepare for the NCP-AIO exam.