Practice for the Red Hat Certified Specialist in OpenShift AI (EX267) exam with 17 exam-style practice questions, instant answer reveals, and concise explanations of every correct answer. Topics include: What is the primary visual interface used by data scientists to track and view pipeline run histories and metrics?. Follow @CertPunch and visit certpunch.com for more certification practice exams and study content.
Prefer hands-on? Take this round as an interactive practice test — answer every question, get instant feedback, and see your score: Start the Red Hat Certified Specialist in OpenShift AI (EX267) practice test →
What you will practice
- What is the primary visual interface used by data scientists to track and view pipeline run histories and met…
- What does the OpenShift AI Dashboard use to determine which notebook images are displayed to users in the wor…
- What action should an administrator execute if a component status in the DataScienceCluster shows 'Degraded'?
- Which command provides detailed validation status information for an active DataScienceCluster configuration…
- Which component handles the automatic orchestration and execution of Data Science Pipelines in OpenShift AI?
- What is the purpose of the CodeFlare stack within Red Hat OpenShift AI?
Answers and explanations
Tap a question to expand the answer and the exam reasoning. Try to commit to your own pick first.
Q1. What is the primary visual interface used by data scientists to track and view pipeline run histories and metrics?
Answer: B. Red Hat OpenShift AI Dashboard
Data scientists use the Red Hat OpenShift AI Dashboard to track pipeline metrics and histories. While Tekton executes the underlying steps, its CLI is not the primary visual interface for this task.
Q2. What does the OpenShift AI Dashboard use to determine which notebook images are displayed to users in the workbench creation screen?
Answer: A. The custom resources of type NotebookImage or ImageStream with specific labels
The dashboard dynamically discovers images using NotebookImage custom resources or labeled ImageStreams. Hardcoded lists are incorrect because the operator queries the cluster API for available images.
Q3. What action should an administrator execute if a component status in the DataScienceCluster shows 'Degraded'?
Answer: C. Review the conditions and logs of the component operator pods to identify the failing sub-resource
Administrators should inspect the conditions and logs of the failing component operator pods. Destructive actions like deleting configurations or wiping persistent volumes are incorrect and will exacerbate the degradation.
Q4. Which command provides detailed validation status information for an active DataScienceCluster configuration deployment?
Answer: A. oc describe datasciencecluster
The oc describe datasciencecluster command displays the validation status and deployment conditions for the cluster. The other options reference unrelated container tools or cluster-wide machine configurations.
Q5. Which component handles the automatic orchestration and execution of Data Science Pipelines in OpenShift AI?
Answer: C. OpenShift Pipelines (Tekton)
Data Science Pipelines rely on OpenShift Pipelines, which is built on Tekton, to orchestrate workflow execution. Jenkins and GitOps do not natively handle the pipeline API calls required by the dashboard.
Q6. What is the purpose of the CodeFlare stack within Red Hat OpenShift AI?
Answer: D. To manage large scale distributed computing and AI workloads
CodeFlare manages and scales large distributed AI workloads across compute resources. It abstracts infrastructure complexity, leaving database migrations and frontend dashboards to other tools.
Q7. What type of resource contains the definition lists of supported hardware accelerators visible in the OpenShift AI cluster settings?
Answer: A. AcceleratorProfile custom resources
AcceleratorProfile custom resources define hardware accelerators like GPUs within OpenShift AI settings. The cluster administrator creates these profiles to expose compatible hardware to data scientists.
Q8. Which tool or backend serves as the foundation for the single-model serving stack in Red Hat OpenShift AI?
Answer: D. KServe
KServe serves as the foundational backend for the single-model serving stack in OpenShift AI. While Seldon Core was used in older Open Data Hub versions, current exams focus entirely on KServe.
Q9. What protocol is used by the Elyra pipeline tool to upload compiled workflows directly to the active OpenShift AI Pipeline Server?
Answer: C. HTTP/HTTPS API requests targeting the pipeline server endpoint
Elyra communicates with the Pipeline Server using HTTP or HTTPS API requests to upload compiled workflows. Git hooks and UDP broadcasts cannot interact with the server's REST API endpoints.
Q10. Which configuration component defines the internal layout structure and default packages of the base Jupyter images provided by Red Hat?
Answer: B. Open Data Hub notebook container manifests
Open Data Hub notebook container manifests define the base image layouts and default packages. Storage profiles and API server configurations handle cluster mechanics, not workbench image contents. Look for notebook manifests.
Q11. Which custom resource configures the global installation settings for Red Hat OpenShift AI Components?
Answer: C. DataScienceCluster
The DataScienceCluster custom resource manages the global configuration for installed OpenShift AI components. The operator processes this resource to determine which features are enabled across the cluster.
Q12. What tool can be deployed alongside OpenShift AI to track model metrics, parameters, and run histories natively across experiments?
Answer: C. MLflow or OpenShift AI tracking component integration
OpenShift AI integrates natively with MLflow to track experiment metrics, parameters, and run histories for data science projects. Prometheus is used for cluster infrastructure metrics rather than machine learning experiment tracking.
Q13. What is the standard format required for exporting a machine learning model to be served by ModelMesh?
Answer: A. ONNX, Pickle, or SavedModel format
ModelMesh supports standard serialized model formats like ONNX, Pickle, and SavedModel to efficiently serve predictions. Raw scripts or dataset files are not directly deployable as inference models.
Q14. When a data scientist spawns a Jupyter notebook instance in OpenShift AI, where are the custom environment variables defined within the project applied?
Answer: A. Directly to the single-user notebook pod configuration
Custom environment variables are injected directly into the configuration of the specific single-user notebook pod. This ensures data scientists have the necessary environment configurations isolated to their active session.
Q15. What type of OpenShift object maps directly to a "Data Science Project" created within the OpenShift AI interface?
Answer: D. A Kubernetes Namespace/Project
A Data Science Project maps directly to a standard Kubernetes Namespace. This provides logical isolation and access control for your workbenches, models, and pipelines within the cluster.
Q16. In OpenShift AI, what mechanism is used to securely store and inject environment variables or database credentials into a notebook?
Answer: D. Data Connections
Data Connections simplify this process by wrapping standard Kubernetes Secrets and ConfigMaps to securely mount credentials as environment variables. Raw ConfigMaps alone are not designed for sensitive password storage.
Q17. Which underlying open-source project forms the basis for OpenShift AI pipelines?
Answer: A. Kubeflow Pipelines
OpenShift AI Pipelines are built directly on top of Kubeflow Pipelines, using Argo Workflows under the hood. Tekton is the native CI/CD tool for OpenShift but not the basis for data science pipelines.
More Red Hat Certified Specialist in OpenShift AI (EX267) drills and other practice exams are on @CertPunch. New rounds drop every few days at certpunch.com.