Analyzing & Optimizing
Performance analysis, monitoring, and optimization strategies
Sample GCP PCA questions with explanations
These six general GCP PCA questions come from the full practice bank; they are not filtered to the selected domain. Review the answers and explanations for cross-domain practice, then choose a focused set above.
You need to orchestrate a complex ML workflow: Data Extraction -> Validation -> Training -> Evaluation -> Deployment. You want a managed, serverless way to define this as a DAG (Directed Acyclic Graph). What tool works natively with Vertex AI?
- ✓ Vertex AI Pipelines (Kubeflow Pipelines SDK).
- ○ Cloud Composer (Airflow).
- ○ Workflows.
- ○ Cloud Batch.
You're conducting a pre-migration security review. Which TWO areas require assessment?
- ✓ Current security controls and their GCP equivalents.
- ✓ Compliance requirements and how GCP addresses them.
- ○ Security can be addressed post-migration.
- ○ No security review needed.
- ○ Only perimeter security matters.
You're designing a mobile backend that needs to support offline-first functionality, real-time synchronization when online, and automatic conflict resolution. Which TWO Google Cloud services best support this requirement?
- ✓ Cloud Firestore with offline persistence.
- ✓ Firebase Realtime Database.
- ○ Cloud SQL with custom sync logic.
- ○ Cloud Storage with versioning.
- ○ BigQuery with batch imports.
Your RAG system retrieves too many irrelevant documents. How can you improve retrieval quality?
- ✓ Tune the similarity threshold and use hybrid search (combining semantic and keyword matching).
- ○ Increase the number of retrieved documents.
- ○ Disable semantic search.
- ○ Use larger embedding dimensions only.
Your GKE HPA is configured to scale at 70% CPU, but pods aren't scaling despite 90% CPU usage. What should you check first?
- ✓ Verify that pods have resource requests defined. HPA requires requests to calculate utilization.
- ○ Increase the cluster size manually.
- ○ Disable cluster autoscaler.
- ○ Change HPA to scale on memory.
You're implementing data governance with Dataplex. Which TWO capabilities does it provide?
- ✓ Data discovery and cataloging across GCP data sources.
- ✓ Data quality validation and profiling.
- ○ Direct data transformation.
- ○ Replacing all existing databases.
- ○ Only works with BigQuery.