Hard1 markMultiple Choice
Domain 4.2: Managing GKE resourcesDomain 4.2GKENode PoolsGPU

GCP ACE · Question 35 · Domain 4.2: Managing GKE resources

Your GKE cluster currently runs on a node pool of e2-standard-2 machines. You are deploying a new machine learning workload that requires GPUs. You cannot attach GPUs to the existing e2 instances.

Which TWO steps should you take to run this new workload on the cluster? (Select TWO)

Answer options:

A.

Edit the existing node pool to change the machine type to n1-standard-4.

B.

Create a new node pool in the existing cluster using a machine type that supports GPUs (e.g., n1-standard-4) and attach GPUs.

C.

Create a completely new GKE cluster just for the GPU workloads.

D.

Use node selectors or node affinities in your pod deployment to schedule the ML pods onto the new GPU node pool.

E.

Run kubectl upgrade nodes --add-gpu.

How to approach this question

Understand how to add different hardware types to an existing GKE cluster.

Full Answer

Create a new node pool in the existing cluster using a machine type that supports GPUs (e.g., n1-standard-4) and attach GPUs., Use node selectors or node affinities in your pod deployment to schedule the ML pods onto the new GPU node pool.
GKE clusters can contain multiple 'node pools', each with different machine types. Since you cannot change the machine type of an existing pool, you must create a new node pool with GPU-compatible instances (like N1 or A2). Then, you use Kubernetes scheduling features (like nodeSelector or nodeAffinity) in your deployment manifest to ensure the ML pods are placed on the new GPU nodes.

Common mistakes

Thinking you must create a whole new cluster, or trying to edit the existing node pool's machine type.

Practice the full GCP Associate Cloud Engineer Practice Exam 4

50 questions · hints · full answers · grading

More questions from this exam