Compute resource management in MLOps - Interactive Code Practice

Practice - 5 Tasks

1fill in blank

easy

Complete the code to request 4 CPUs for a Kubernetes pod.

MLOps

resources:
  requests:
    cpu: [1]

Drag options to blanks, or click blank then click option'

A4

B2

C8

D1

Attempts:

3 left

💡 Hint

Common Mistakes

Using memory units instead of CPU

Specifying CPU as a string with units like '4m'

Setting cpu: 4 requests 4 CPUs for the pod.

2fill in blank

medium

Complete the command to check current GPU usage on a node.

MLOps

nvidia-smi [1]

Drag options to blanks, or click blank then click option'

A--query-gpu=memory.used

B--list-gpus

C--help

D--version

Attempts:

3 left

💡 Hint

Common Mistakes

Using --list-gpus which only lists GPUs

Using --version which shows version info

The --query-gpu=memory.used option shows GPU memory usage.

3fill in blank

hard

Fix the error in the YAML to limit memory to 8Gi.

MLOps

resources:
  limits:
    memory: [1]

Drag options to blanks, or click blank then click option'

A8000Mi

B8GB

C8Gi

D8g

Attempts:

3 left

💡 Hint

Common Mistakes

Using 8GB which is invalid

Using lowercase g instead of Gi

Memory limits in Kubernetes use binary SI units like Gi, so 8Gi is correct.

4fill in blank

hard

Fill both blanks to create a resource request for 2 CPUs and 4Gi memory.

MLOps

resources:
  requests:
    cpu: [1]
    memory: [2]

Drag options to blanks, or click blank then click option'

A2

B4Gi

C8Gi

D1

Attempts:

3 left

💡 Hint

Common Mistakes

Mixing CPU and memory units

Using incorrect memory units like GB

Requesting 2 CPUs and 4Gi memory uses cpu: 2 and memory: 4Gi.

5fill in blank

hard

Fill all three blanks to define a pod with 1 CPU request, 2Gi memory request, and 1 GPU request.

MLOps

resources:
  requests:
    cpu: [1]
    memory: [2]
    nvidia.com/gpu: [3]

Drag options to blanks, or click blank then click option'

A1

B2Gi

D4Gi

Attempts:

3 left

💡 Hint

Common Mistakes

Using memory units for CPU or GPU

Incorrect GPU key name

The pod requests 1 CPU, 2Gi memory, and 1 GPU using the correct keys and values.