A team uses Kubeflow to run ML pipelines on Kubernetes. They need to reuse a preprocessing step across multiple pipelines. What is the BEST way to achieve this in Kubeflow?

Use a Kubernetes Job for each pipeline

Run the step as a separate service and call it from pipelines

Duplicate the step code in each pipeline

Create a reusable pipeline component and reference it in multiple pipelines

A company uses Azure Machine Learning to deploy a model. They need to update the model without downtime and be able to roll back quickly if the new version performs poorly. What should they configure?

Use Azure Container Instances to run the model

A single deployment that is overwritten with the new model

A new endpoint for each model version

Multiple deployments behind the same endpoint with traffic splitting

A team uses Azure Machine Learning to deploy a model as a real-time endpoint. They need to support a sudden spike in traffic during a product launch, but they want to minimize cost during normal periods. The model is containerized. What is the BEST deployment approach?

Deploy to Azure Functions with a consumption plan

Deploy to a single Azure ML compute instance

Deploy to Azure Container Instances (ACI) with a fixed number of containers

Deploy to Azure Kubernetes Service (AKS) with cluster autoscaling and multiple model deployments behind one endpoint

A data scientist builds a Docker image for a model. The image must be small and start quickly. Which Dockerfile practice BEST supports this goal?

Use a slim base image and install only required dependencies

Use a large base image with many pre-installed tools

Copy the entire project directory including data files

Install all available Python packages to ensure compatibility

CertNexusCertified Data Science Practitioner (CDSP)

Domain 6Objective 2

Deploy and Manage Models CERTIFIED-DATA-SCIENCE-PRACTITIONER Practice Questions (Page 6)

Part of the 6.0 Operationalizing the pipeline domain, which accounts for 5-8% of the CERTIFIED-DATA-SCIENCE-PRACTITIONER exam.

36questions here

8free pages

8concepts

5-8%of the exam

Questions 26–30

26application · medium
A team uses Kubeflow to run ML pipelines on Kubernetes. They need to reuse a preprocessing step across multiple pipelines. What is the BEST way to achieve this in Kubeflow?
Select an answer first
27application · medium
A company uses Azure Machine Learning to deploy a model. They need to update the model without downtime and be able to roll back quickly if the new version performs poorly. What should they configure?
Select an answer first
28expert · hard
A team uses Azure Machine Learning to deploy a model as a real-time endpoint. They need to support a sudden spike in traffic during a product launch, but they want to minimize cost during normal periods. The model is containerized. What is the BEST deployment approach?
Select an answer first
29foundation · easy
A team wants to automate the build, test, and deployment of their Azure Machine Learning models whenever code is pushed to a repository. Which Azure service is commonly used to set up this CI/CD pipeline?
Select an answer first
30application · medium
A data scientist builds a Docker image for a model. The image must be small and start quickly. Which Dockerfile practice BEST supports this goal?
Select an answer first

Finished these 5 questions?

Review the revealed explanations, or continue through the curriculum.

Next 5 questions →Back to CERTIFIED-DATA-SCIENCE-PRACTITIONER

Free Basic Practice is a study aid with revealable answers — not a scored exam. Examers.io is independent and not affiliated with or endorsed by CertNexus. “CERTIFIED-DATA-SCIENCE-PRACTITIONER” is a trademark of its owner, used for identification only.