How to reduce Vertex AI Model Monitoring costs for high traffic?

You recently deployed a model to a Vertex AI endpoint. Your data drifts frequently, so you have enabled request-response logging and created a Vertex AI Model Monitoring job. You have observed that your model is receiving higher traffic than expected. You need to reduce the model monitoring cost while continuing to quickly detect drift. What should you do?

Replace the monitoring job with a DataFlow pipeline that uses TensorFlow Data Validation (TFDV)
Replace the monitoring job with a custom SQL script to calculate statistics on the features and predictions in BigQuery
Decrease the sample_rate parameter in the RandomSampleConfig of the monitoring job Source Reference Answer
Increase the monitor_interval parameter in the ScheduleConfig of the monitoring job

Community Votes

100%

100% of anonymous learners picked answer C. Votes are pick records left by other test-takers — they are not the verified answer.

Community Insight

This question tests the distinction between monitoring volume and frequency, requiring you to reduce cost via sampling rather than delaying checks.

To reduce Vertex AI Model Monitoring costs under high traffic without delaying drift detection, adjusting the sampling configuration is the optimal solution. The community confirms that decreasing the sample_rate effectively lowers expenses while maintaining the frequency of analysis.

Selecting Option D is a frequent error because it lowers costs but increases the time to detect drift, violating the requirement for quick detection.

Community Discussion (6 comments)

LaxmanTiwari 👍 1

fitri001 2 months, 1 week ago A. DataFlow pipeline with TFDV: While DataFlow pipelines with TFDV can be used for data validation, they require additional development and management overhead compared to simply adjusting the Vertex AI Model Monitoring job configuration. B. Custom SQL script: Custom SQL scripts might not be as efficient or maintainable as the built-in Vertex AI Model Monitoring features. Additionally, it would require manually calculating drift metrics, which can be error-prone. D. Increase monitor_interval: Increasing the monitoring interval reduces the frequency of monitoring checks, potentially delaying drift detection. This is not ideal if data drifts frequently.

fitri001 👍 3 Selected: C

Reduced Monitoring Overhead: By decreasing the sample_rate, you instruct Vertex AI Model Monitoring to analyze a smaller percentage of incoming requests. This directly reduces the billing cost associated with monitoring. Fast Drift Detection: A well-chosen sampling rate can still provide enough data to capture significant data drift. Monitoring a smaller sample shouldn't significantly impact your ability to detect drift if it's happening rapidly.

Carlose2108 👍 1 Selected: C

I went with C.

ddogg 👍 2 Selected: C

C as the sample size will be relative to the traffic and also reduce costs.

b1a8fae 👍 1 Selected: C

C. https://cloud.google.com/vertex-ai/docs/model-monitoring/overview#considerations

pikachu007 👍 2 Selected: C

The answer is C, simplest and does not affect the time it takes to detect the drift

Comments & Corrections

No comments yet — spotted an error or have a note? Share it below.

Expert Analysis

Why the Answer Is Correct

Decreasing the sample_rate in RandomSampleConfig directly reduces the number of requests analyzed for statistical drift, thereby lowering storage and compute costs. Since the monitoring schedule remains unchanged, the system continues to run analysis jobs at the same frequency. This ensures that drift is detected quickly while significantly reducing the overhead associated with high traffic volumes.

Why the Other Options Are Wrong

Option A suggests replacing the managed service with a DataFlow pipeline, which introduces unnecessary engineering complexity and infrastructure management. Option B proposes a custom SQL script, which is manual, lacks automated alerting, and is less maintainable than the native feature. Option D increases the monitor_interval, which reduces cost but also delays the time it takes to identify drift, failing the specific requirement to detect it quickly.

Community Comment Notes

Community members strongly support Option C, noting it is the simplest way to manage costs without impacting detection speed. One commenter explained that a reduced sample size is sufficient to capture significant drift trends. Another user cited official Google Cloud documentation to validate that adjusting the sampling configuration is the recommended approach for cost optimization.

Official Reference

https://cloud.google.com/vertex-ai/docs/model-monitoring/overview#considerations

Exam Strategy

When faced with cost optimization questions involving monitoring or logging, prioritize reducing data volume (sampling) over increasing time intervals. This ensures you meet performance requirements for detection speed while adhering to budget constraints.

Related Analysis

Practice All PMLE Questions

Access 65 questions with complete answers and detailed explanations.

View Full PMLE Practice Test →

← Back to PMLE Study Guide