mediumMultiple Choice
PDE Practice Question: A company deploys a machine learning model to…
A company deploys a machine learning model to Vertex AI for real-time predictions. After deployment, they notice that prediction latency spikes during peak traffic hours. Which approach should they take to reduce latency without sacrificing accuracy?
⚠ Common exam trap
Google Cloud often tests the misconception that reducing features or using batch prediction is the primary way to reduce latency, but the real exam trap is that candidates overlook the need to maintain real-time capability and accuracy, and instead choose a solution that changes the model or prediction mode rather than scaling infrastructure.
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
✓
Configure auto-scaling with higher min and max instances
Configuring auto-scaling with higher min and max instances ensures that Vertex AI has sufficient pre-warmed replicas to handle traffic spikes without cold-start latency. This approach maintains model accuracy because it does not alter the model architecture or inference logic, only the infrastructure capacity.
Answer analysis
Option-by-option breakdown
For each option: why learners choose it and why it is or isn't the right answer here.
- ✓
Configure auto-scaling with higher min and max instances
Why this is correct
Auto-scaling with higher minimum and maximum instances adds serving capacity during peak traffic, absorbing load and cutting latency while the same model preserves accuracy. This satisfies the stem's constraint of reducing latency without sacrificing accuracy.
- ✗
Reduce the number of input features
Why it's wrong here
Dropping input features changes the model's inputs, degrading prediction quality and so sacrificing accuracy, which the scenario forbids. It is tempting because fewer features reduce per-request computation, and would be correct only when retraining shows the removed features contribute negligibly to predictions.
- ✗
Switch from online to batch prediction
Why it's wrong here
Batch prediction processes accumulated requests offline, so it cannot serve real-time responses at all, eliminating the latency requirement rather than reducing it. It is tempting because batch jobs avoid per-request overhead, and would be correct for scheduled scoring where throughput matters and immediate answers are unnecessary.
- ✗
Use a larger machine type for the model
Why it's wrong here
A larger machine type adds CPU, memory or accelerators, which does not address the concurrency bottleneck causing peak-hour latency spikes unless the model is genuinely resource-starved. It is tempting because scaling up is a familiar remedy, and would be correct if profiling showed the instance was saturated on compute.
Go deeper
Related to this question
About these practice questions
One of 747 original PDE practice questions on Courseiva, each with a full explanation and wrong-answer analysis — not exam dumps or protected exam content. Learn why practice questions differ from exam dumps →
JA
Written by Johnson Ajibi, MSc IT Security
Senior Network & Security Engineer · founder of Courseiva
This PDE practice question is part of Courseiva's free Google Cloud certification practice question bank. Courseiva provides original exam-style practice questions with explanations, topic-based practice, mock exams, readiness tracking, and study analytics to help learners prepare for the PDE exam.