PMLE Practice Question: Your team manages multiple ML models on Vertex AI

Your team manages multiple ML models on Vertex AI. You need to implement a centralized monitoring solution to track model performance over time. Which TWO approaches should you consider? (Choose two.)

⚠ Common exam trap

Watch out — candidates often confuse logging (Option A) or cost tracking (Option E) with performance monitoring, or mistakenly think version control (Option B) is part of monitoring, when the question specifically asks for centralized monitoring of model performance over time.

Answer choices

Why each option matters

Answer the question above first, then reveal the full breakdown to understand why each option is right or wrong.

Correct answer & explanation

✓

Create Cloud Monitoring dashboards and alerts based on Vertex AI metrics.

Option C is correct because Cloud Monitoring natively ingests Vertex AI metrics (e.g., prediction counts, latency, error rates, and Model Monitoring anomaly signals) and lets you build centralized dashboards and alerting policies across all deployed models in one place. Option D is correct because Vertex AI Model Monitoring is the purpose-built service that continuously computes training-serving skew and prediction drift against a baseline for each deployed model, emitting metrics and alerts that feed the centralized monitoring view. Option A is not the intended answer because storing prediction logs in BigQuery is a data-warehouse analysis pattern, not a centralized monitoring solution with dashboards and alerts. Option B is incorrect because Cloud Source Repositories is a Git version-control service for source code, not a runtime performance monitoring tool. Option E is incorrect because Cloud Billing budgets track spend, not model performance metrics such as skew or drift.

Answer analysis

Option-by-option breakdown

For each option: why learners choose it and why it is or isn't the right answer here.

✗
Store all prediction logs in BigQuery and analyze using SQL.
Why it's wrong here
BigQuery stores prediction logs for SQL analysis, but logging alone provides no drift detection, alerting, or scheduled evaluation. Vertex AI Model Monitoring computes skew and drift against training baselines and raises alerts. BigQuery is the right choice for ad-hoc analysis or custom dashboards, not automated monitoring.
✗
Use Cloud Source Repositories to track model code versions.
Why it's wrong here
Cloud Source Repositories stores Git source code, giving version history but no runtime telemetry. Tracking model performance over time needs prediction outputs and ground-truth labels captured at serving time, which source control never records. It would be correct for auditing code changes or rolling back a training script.
✓
Create Cloud Monitoring dashboards and alerts based on Vertex AI metrics.
Why this is correct
Cloud Monitoring natively ingests Vertex AI's built-in metrics, such as prediction drift and training-serving skew, so dashboards and alerts can be centralised across every deployed model without custom instrumentation. This directly satisfies the requirement for a centralised monitoring solution tracking performance over time.
✓
Use Vertex AI Model Monitoring to detect training-serving skew and feature drift for each model.
Why this is correct
Vertex AI Model Monitoring natively computes training-serving skew and feature drift against a baseline for deployed models, satisfying the requirement for centralised, per-model performance tracking without bespoke pipelines. It alerts when distributions deviate, giving the team ongoing visibility across all models.
✗
Enable Cloud Billing budgets to track cost per model.
Why it's wrong here
Cloud Billing budgets track spend against thresholds, reporting cost per model rather than predictive quality. Performance monitoring requires comparing served predictions with training baselines or labels to detect drift. Budgets would be the correct choice for controlling Vertex AI expenditure or alerting finance when consumption exceeds a set amount.

Go deeper

PMLE Practice Question: Your team manages multiple ML models on Vertex AI

Why each option matters

Option-by-option breakdown

Related to this question