PMLE Monitoring ML Solutions Practice Question

A logistics company has a Vertex AI Model Monitoring job that detects feature drift on a deployed route optimization model. The model uses 50 features, and the monitoring job is configured to monitor all features. The MLOps team notices that the monitoring job is incurring high costs and taking a long time to complete. They want to reduce monitoring overhead while still detecting significant drift on the most important features. What should they do?

⚠ Common exam trap

The trap here is focusing on monitoring frequency or sampling rate as the primary cost driver, when the number of monitored features is often the dominant factor in per-run cost and runtime.

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

✓

Select a subset of features to monitor based on feature importance or business relevance.

Monitoring all 50 features increases cost and runtime. To reduce overhead while still detecting significant drift, the team should select a subset of features to monitor based on importance or business relevance. This reduces the computational load per run. Reducing frequency lowers total runs but not per-run cost. Increasing sampling rate would increase cost. Disabling numerical features arbitrarily could miss critical 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.

✓
Select a subset of features to monitor based on feature importance or business relevance.
Why this is correct
Monitoring fewer features reduces the computational load and cost per run. By focusing on the most important features, the team can still detect significant drift while lowering overhead. Vertex AI Model Monitoring allows you to specify which features to monitor. This targeted approach is the most effective way to reduce cost and runtime without sacrificing detection of critical drift.
✗
Disable drift detection for numerical features and only monitor categorical features.
Why it's wrong here
Disabling numerical feature drift may reduce cost but could miss critical drift in numerical features, which are likely important for route optimization. This is an arbitrary restriction not based on feature importance. The team should selectively monitor features based on relevance, not by data type. This option sacrifices detection capability without a justified reason.
✗
Increase the sampling rate to reduce the volume of prediction requests analyzed.
Why it's wrong here
Increasing the sampling rate means analyzing a larger proportion of requests, which would increase cost and runtime. To reduce overhead, you would decrease the sampling rate. However, even with lower sampling, analyzing all 50 features still incurs significant cost. The better approach is to reduce the number of features monitored. This option would worsen the problem.
✗
Reduce the monitoring frequency from daily to weekly.
Why it's wrong here
Reducing frequency lowers the number of monitoring runs but does not reduce the cost per run, which is driven by the number of features analyzed. The job will still process all 50 features each time, so per-run cost remains high. While overall cost may decrease due to fewer runs, the long runtime per run persists. This option does not address the core issue of analyzing too many features.

Go deeper

PMLE Monitoring ML Solutions Practice Question

Why each option matters

Option-by-option breakdown

Related to this question