Question 307 of 506
Monitoring ML solutionsmediumMultiple ChoiceObjective-mapped

Quick Answer

The answer is to increase the batch size to 64 and the batch timeout to 200ms. This is correct because the core problem is low GPU utilization at 10% despite rising latency, which indicates the GPU is underutilized and the bottleneck lies in inefficient batching. By increasing both the batch size and timeout, TensorFlow Serving can accumulate more inference requests per batch, allowing the GPU to process them in parallel and dramatically improving throughput while reducing per-request latency. On the Google Professional Machine Learning Engineer exam, this scenario tests your understanding of optimizing batching for Vertex AI GPU inference, often appearing as a trap where candidates mistakenly adjust autoscaling or CPU targets instead of tuning the serving batching parameters. A common memory tip is "low GPU utilization + high latency = batch too small or timeout too short," so always look to increase both values to better leverage GPU parallelism.

PMLE Monitoring ML solutions Practice Question

This PMLE practice question tests your understanding of monitoring ml solutions. The scenario asks you to isolate a root cause — eliminate options that address a different problem before choosing. After answering, compare your reasoning against the explanation and wrong-answer breakdown below. Once you have made your selection, read the full explanation to reinforce the concept and understand why each distractor is designed to mislead on exam day.

You are an ML engineer at a logistics company. You have deployed a deep learning model on Vertex AI Endpoints using a custom container with GPU acceleration. The model predicts delivery times based on route features. After one week, you notice that the endpoint's GPU utilization is consistently at 10%, but the prediction latency has increased by 50%. The number of prediction requests per second has remained stable. You check the container logs and see no errors. The model is served using TensorFlow Serving with batching enabled (batch size: 32, batch timeout: 100ms). The custom container uses a single NVIDIA T4 GPU. You have also set the Vertex AI endpoint to use autoscaling with minReplicaCount: 1 and maxReplicaCount: 5, and the CPU utilization target is 60%. Which action should you take to reduce latency?

Question 1mediummultiple choice
Review the full routing breakdown →

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

Increase the batch size to 64 and batch timeout to 200ms to improve GPU utilization.

The core issue is low GPU utilization (10%) despite increased latency, indicating that the GPU is underutilized and the bottleneck is likely in batching or data pipeline overhead. Increasing the batch size to 64 and batch timeout to 200ms allows TensorFlow Serving to accumulate more requests per batch, improving GPU throughput and reducing per-request latency by better leveraging GPU parallelism. This directly addresses the mismatch between low GPU utilization and high latency.

Key principle: Answer the scenario, not the keyword: identify the specific constraint before choosing the most familiar-sounding option.

Answer analysis

Option-by-option breakdown

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

  • Increase the minReplicaCount to 3 to handle requests in parallel.

    Why it's wrong here

    Adding more replicas does not address the low GPU utilization per replica; it may spread the load but each replica still underutilizes GPU.

  • Reduce the CPU utilization target to 40% to trigger more aggressive autoscaling.

    Why it's wrong here

    CPU utilization is not the bottleneck; lowering the target would add more replicas, not improve GPU efficiency.

  • Quantize the model to FP16 to reduce compute time per inference.

    Why it's wrong here

    Quantization can help, but the primary issue is low GPU utilization due to small batches; optimizing batching should be the first step.

  • Increase the batch size to 64 and batch timeout to 200ms to improve GPU utilization.

    Why this is correct

    Larger batch sizes allow the GPU to process more data per inference, increasing throughput and reducing per-request latency once the batch fills up.

    Related concept

    Read the scenario before looking for a memorised answer.

Common exam traps

Common exam trap: answer the scenario, not the keyword

The trap here is that candidates focus on scaling or model optimization (A, B, C) without recognizing that low GPU utilization with high latency is a classic sign of inefficient batching, not insufficient compute or replicas.

Detailed technical explanation

How to think about this question

TensorFlow Serving's batching mechanism uses a dynamic scheduler that waits for either a full batch or the batch timeout to elapse. With a batch size of 32 and timeout of 100ms, the GPU may be processing small batches frequently, leading to low utilization. Increasing both parameters allows the scheduler to form larger batches, improving GPU occupancy and reducing the number of kernel launches, which is a common optimization for latency-sensitive serving with NVIDIA T4 GPUs. In practice, monitoring GPU utilization alongside latency helps distinguish between compute-bound and I/O-bound bottlenecks.

KKey Concepts to Remember

  • Read the scenario before looking for a memorised answer.
  • Find the constraint that changes the correct option.
  • Eliminate answers that are true in general but not in this case.

TExam Day Tips

  • Watch for words such as best, first, most likely and least administrative effort.
  • Review why wrong options are wrong, not only why the correct option is correct.

Key takeaway

Answer the scenario, not the keyword: identify the specific constraint before choosing the most familiar-sounding option.

Real-world example

How this comes up in practice

An e-commerce site experiences heavy traffic on Black Friday and near-zero traffic during off-peak weeks. Rather than provisioning permanent large VMs, the team uses auto-scaling groups that add capacity automatically under load and reduce it overnight. Questions like this test whether you understand elasticity, availability zones, and cloud compute scaling patterns.

What to study next

Got this wrong? Here's your next step.

Identify which exam domain this question belongs to, review the core concept, then practise similar questions from the same domain.

Related practice questions

Related PMLE practice-question pages

Use these pages to review the topic behind this question. This is how one missed question becomes focused revision.

Practice this exam

Start a free PMLE practice session

Short sessions build daily habit. Longer sessions build exam-day stamina. Try a timed session to simulate real conditions.

FAQ

Questions learners often ask

What does this PMLE question test?

Monitoring ML solutions — This question tests Monitoring ML solutions — Read the scenario before looking for a memorised answer..

What is the correct answer to this question?

The correct answer is: Increase the batch size to 64 and batch timeout to 200ms to improve GPU utilization. — The core issue is low GPU utilization (10%) despite increased latency, indicating that the GPU is underutilized and the bottleneck is likely in batching or data pipeline overhead. Increasing the batch size to 64 and batch timeout to 200ms allows TensorFlow Serving to accumulate more requests per batch, improving GPU throughput and reducing per-request latency by better leveraging GPU parallelism. This directly addresses the mismatch between low GPU utilization and high latency.

What should I do if I get this PMLE question wrong?

Identify which exam domain this question belongs to, review the core concept, then practise similar questions from the same domain.

What is the key concept behind this question?

Read the scenario before looking for a memorised answer.

About these practice questions

Courseiva creates original exam-style practice questions with explanations and wrong-answer analysis. It does not publish real exam questions, exam dumps, or protected exam content. Learn why practice questions differ from exam dumps →

How Courseiva writes practice questions · Editorial policy

Last reviewed: Jun 11, 2026

Question Discussion

Share a tip, memory trick, or ask about the reasoning behind this question. Do not post real exam questions, leaked content, braindumps, or copyrighted exam material. Comments are moderated and may be removed without notice.

Loading comments…

Sign in to join the discussion.

This PMLE 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 PMLE exam.