Kubeflow Pipelines overview in MLOps - Practice Problems & Coding Challenges

Challenge - 5 Problems

🧠 Conceptual

intermediate

What is the primary purpose of Kubeflow Pipelines?

Choose the best description of what Kubeflow Pipelines is mainly used for.

A To monitor server health and resource usage in Kubernetes clusters.

B To automate and manage machine learning workflows on Kubernetes.

C To deploy web applications using container orchestration.

D To store large datasets for machine learning training.

💡 Hint

Think about what Kubeflow Pipelines helps automate in ML projects.

Kubeflow Pipelines is designed to create, deploy, and manage end-to-end machine learning workflows on Kubernetes.

💻 Command Output

intermediate

Output of listing Kubeflow Pipelines runs

What is the expected output of the command kubectl get runs -n kubeflow if there are two pipeline runs named 'train-model' and 'data-prep'?

MLOps

kubectl get runs -n kubeflow

A

NAME          STATUS    AGE
train-model   Succeeded  2h
data-prep    Running    30m

B

NAME          STATUS    AGE
train-model   Running    2h
data-prep    Succeeded  30m

C

NAME          STATUS    AGE
train-model   Failed     2h
data-prep    Pending    30m

D

NAME          STATUS    AGE
train-model   Succeeded  2h
 data-prep    Running    30m

💡 Hint

Look carefully at the status and spacing in the output.

The output lists pipeline runs with their names, statuses, and ages. 'train-model' succeeded 2 hours ago, 'data-prep' is currently running for 30 minutes.

❓ Configuration

advanced

Correct YAML snippet for a Kubeflow Pipeline component

Which YAML snippet correctly defines a Kubeflow Pipeline component that runs a Python script with an input parameter?

A

apiVersion: v1
kind: Component
metadata:
  name: example-component
spec:
  implementation:
    container:
      image: python:3.8
      command: ["python", "script.py", "--input", "{{inputs.parameters.input_param}}"]
  inputs:
    - name: input_param

B

apiVersion: v1
kind: Component
metadata:
  name: example-component
spec:
  implementation:
    container:
      image: python:3.8
      command: ["python", "script.py", "--input", {input_param}]
  inputs:
    - name: input_param

C

apiVersion: v1
kind: Component
metadata:
  name: example-component
spec:
  implementation:
    container:
      image: python:3.8
      command: ["python", "script.py", "--input", input_param]
  inputs:
    - name: input_param

D

apiVersion: v1
kind: Component
metadata:
  name: example-component
spec:
  implementation:
    container:
      image: python:3.8
      command: ["python", "script.py", "--input", "$input_param"]
  inputs:
    - name: input_param

💡 Hint

Look for the correct syntax to reference input parameters in Kubeflow component YAML.

Kubeflow Pipelines uses the syntax {{inputs.parameters.}} to reference input parameters inside container commands.

🔀 Workflow

advanced

Order of steps in a Kubeflow Pipeline workflow

Arrange the following steps in the correct order for a typical Kubeflow Pipeline workflow:

A 3,1,2,4

B 2,1,3,4

C 1,3,2,4

D 1,2,3,4

💡 Hint

Think about how you build and run a pipeline step by step.

First, you define components, then compile the pipeline, upload it, and finally run and monitor it.

❓ Troubleshoot

expert

Identifying cause of pipeline run failure

A Kubeflow Pipeline run fails immediately with the error message: "Failed to pull image 'myregistry/myimage:latest': unauthorized: authentication required". What is the most likely cause?

A The container image does not exist in the registry.

B The pipeline YAML file has syntax errors causing the failure.

C The Kubernetes cluster nodes lack permission to access the private container registry.

D The pipeline components are missing required input parameters.

💡 Hint

Think about what "unauthorized: authentication required" means when pulling images.

This error usually means the cluster cannot authenticate to the private container registry to pull the image.