PMLE Practice Test 22 — 15 Questions

Question 1

You are designing a distributed training job for a very large neural network that does not fit on a single machine. You need to split the model across multiple devices. Which TWO techniques can you use?

Choose an answer to begin — your selection is scored in the full session.

Accepted Answer

Pipeline parallelism. Pipeline parallelism splits layers across devices; model parallelism (operator-level) splits individual operations. Data parallelism replicates the model. ParameterServer is a form of distributed training, not model splitting.

Answer

ParameterServerStrategy

Answer

Data parallelism with MirroredStrategy

Answer

MultiWorkerMirroredStrategy

Question 2

You need to run a custom training job on Vertex AI using a pre-built container for scikit-learn. Which container image should you specify?

Accepted Answer

us-docker.pkg.dev/vertex-ai/training/scikit-learn-cpu.0.23-0. Vertex AI provides pre-built containers for scikit-learn with the prefix 'us-docker.pkg.dev/vertex-ai/training/scikit-learn-cpu.0.23-0'. The others are for different frameworks.

Answer

us-docker.pkg.dev/vertex-ai/training/pytorch-gpu.1-9

Answer

us-docker.pkg.dev/vertex-ai/training/xgboost-cpu.1-3

Answer

us-docker.pkg.dev/vertex-ai/training/tf-cpu.2-6

Question 3

You are deploying a PyTorch model on Vertex AI and want to use NVIDIA Triton Inference Server for optimal performance. You have built a custom container with Triton. Which serving configuration should you use?

Accepted Answer

Upload your Triton container to Container Registry and specify it as the prediction container in Vertex AI Model.. Vertex AI supports custom containers for prediction. You can bring your own container with Triton installed, and Vertex AI will deploy it to the endpoint.

Answer

Deploy the model on GKE with Triton and expose via Istio.

Answer

Use the prebuilt Vertex AI PyTorch prediction container and set environment variables to enable Triton.

Answer

Use Vertex AI Model Optimization to automatically convert the model to TensorRT and deploy with built-in server.

Question 4

A team wants to perform hyperparameter tuning on a Vertex AI custom training job with 100 trials. They require an algorithm that efficiently explores the search space by learning from previous trials. Which algorithm should they select in the study configuration?

Accepted Answer

ALGORITHM_UNSPECIFIED (defaults to Bayesian optimization). Vertex AI Vizier provides Bayesian optimization as a default algorithm that builds a probabilistic model of the objective function and selects hyperparameters based on expected improvement. This is more efficient than grid or random search for most scenarios. The ALGORITHM_UNSPECIFIED defaults to Bayesian optimization.

Answer

RANDOM_SEARCH

Answer

HYPERBAND

Answer

GRID_SEARCH

Question 5

A team is fine-tuning a large language model (LLaMA 2) using Vertex AI with a custom container on a multi-node GPU cluster. They need to implement model parallelism to fit the model across multiple GPUs because it does not fit into a single GPU memory. Which distributed training strategy should they use?

Accepted Answer

Implement pipeline parallelism by manually splitting the model layers across GPUs and using a framework like PyTorch's RPC or Megatron-LM. Model parallelism, specifically pipeline parallelism, splits the model layers across devices. For large models that don't fit on one GPU, this is necessary. Data parallelism (even with ZeRO) still requires the full model on each device. Vertex AI does not natively support model parallelism; users must configure it manually using frameworks like Megatron-LM or DeepSpeed.

Answer

Use Vertex AI Hyperparameter Tuning to find optimal model partitioning

Answer

Use tf.distribute.MirroredStrategy across all GPUs

Answer

Use Vertex AI distributed training with TF_CONFIG to set up multi-worker mirrored strategy and rely on XLA to partition the model

Question 6

You are fine-tuning a BERT model from Hugging Face Transformers on Vertex AI. You want to minimise cost for a short experiment. Which compute configuration should you use?

Accepted Answer

A custom training job with a single NVIDIA T4 GPU using spot VMs. Spot VMs provide up to 60-90% discount compared to regular VMs. Since the experiment is short, preemption risk is low. Custom TPU pods are expensive and overkill; T4 GPUs are cheaper but spot VMs are the most cost-effective.

Answer

A custom training job with a TPU v3-8 pod

Answer

A custom training job with 8 NVIDIA V100 GPUs using regular VMs

Answer

A standard n1-highmem-8 machine with no accelerator

Question 7

An ML team is optimizing an inference model for deployment on edge devices. They need to reduce the model size and improve latency while maintaining accuracy as much as possible. Which two techniques should they use? (Choose TWO.)

Accepted Answer

Post-training quantization to INT8.. Post-training quantization to INT8 reduces model size by converting 32-bit floating-point weights and activations to 8-bit integers, which also speeds up inference on edge devices with integer-optimized hardware. This technique typically maintains accuracy within 1-2% of the original model while significantly lowering memory footprint and latency.

Answer

Use a larger pre-trained model as a starting point.

Answer

Use half-precision (FP16) instead of INT8.

Answer

Increase the number of layers in the model.

Question 8

You need to run a distributed training job on Vertex AI using TensorFlow with MirroredStrategy on a single machine with 4 GPUs. Which training configuration should you use?

Accepted Answer

Use MirroredStrategy with a single workerPoolSpec containing a machine_type with 4 GPUs. For single-machine multi-worker training with multiple GPUs, TensorFlow's MirroredStrategy is appropriate. The workerPoolSpec should have a single worker pool with a machine type that has multiple GPUs.

Answer

Use MultiWorkerMirroredStrategy with multiple workerPools

Answer

Use MirroredStrategy with two workerPoolSpecs, each with 2 GPUs

Answer

Use ParameterServerStrategy with a chief and a parameter server

Question 9

A data science team is building a feature engineering pipeline that processes large-scale data from BigQuery daily. They need to compute aggregate features and store the results in Vertex AI Feature Store for both online serving and offline training. Which Google Cloud service is best suited for this batch computation?

Accepted Answer

Dataflow. Dataflow is ideal for batch processing large datasets from BigQuery with Apache Beam. It can write directly to Feature Store's API. Cloud Functions is event-driven and not for heavy batch. Dataproc is for Spark/Hadoop, not as efficient for Beam. Cloud Composer is an orchestrator, not executor.

Answer

Cloud Composer

Answer

Dataproc

Answer

Cloud Functions

Question 10

Your Vertex AI endpoint is experiencing high latency during traffic spikes. You have set maxReplicas=10 and minReplicas=2. The CPU utilisation target is 60%. During spikes, the endpoint never scales beyond 4 replicas. What is the most likely reason?

Accepted Answer

The maxReplicas limit is set to 10, but the cooldown period is preventing rapid scaling.. The default cooldown period (usually 120 seconds) prevents rapid scaling. If traffic spikes are very short, the endpoint may not trigger a scale-up because the utilisation spike doesn't persist long enough.

Answer

You need to enable GPU acceleration.

Answer

The endpoint is using a legacy model framework.

Answer

The machine type is too small.

Question 11

An ML team has set up automated retraining triggered by Cloud Monitoring alerts. When a feature drift alert fires, a Cloud Function publishes to Pub/Sub, which triggers a Vertex AI Pipeline. However, the retraining pipeline is failing because the training data is not updated. What is the most likely cause?

Accepted Answer

The training data in the pipeline input is stale or not refreshed. Cloud Monitoring alerts can trigger retraining, but the pipeline typically expects the most recent data. If the training data is not refreshed, the pipeline may fail or produce a stale model.

Answer

The Cloud Function does not have permission to start the pipeline

Answer

The Pub/Sub topic is incorrectly configured

Answer

The model endpoint is overloaded

Question 12

A model deployed on a Vertex AI Endpoint uses an image model with XRAI explainability. The team notices that the prediction distributions are shifting over time. They want to monitor prediction drift. However, the explainability feature is not enabled. What must the engineer do to enable monitoring prediction drift?

Accepted Answer

Configure Vertex AI Model Monitoring to monitor prediction drift. Prediction drift monitoring is part of Vertex AI Model Monitoring and does not require explainability to be enabled. It can be configured independently.

Answer

Re-deploy the model with a sampling rate of 100%

Answer

Enable Vertex AI Explainability with XRAI on the endpoint deployment

Answer

Enable request/response logging to BigQuery and build custom drift detection

Question 13

A company wants to cache predictions for identical requests to reduce latency and cost. They use Vertex AI Prediction with a custom container. Which GCP service should they use to implement prediction caching?

Accepted Answer

Cloud Memorystore for Redis. Cloud Memorystore (Redis) is ideal for caching because it provides low-latency key-value storage. The prediction request can be hashed to create a key.

Answer

Cloud Bigtable

Answer

Cloud Storage

Answer

Cloud Firestore

Question 14

A company wants to implement a retraining trigger for their ML model. They have set up Cloud Monitoring alerts that fire when drift exceeds a threshold. What should be the target of the alert to automatically start a Vertex AI Pipeline for retraining?

Accepted Answer

Pub/Sub topic. Cloud Monitoring alerts can send notifications to Pub/Sub topics. A Pub/Sub message can then trigger a Cloud Function that starts a Vertex AI Pipeline run.

Answer

Vertex AI Model Registry

Answer

Cloud Storage bucket

Answer

Cloud Functions HTTP trigger

Question 15

A team is designing a ML pipeline that includes training, evaluation, and conditional deployment. They want to use Vertex AI Pipelines. Which THREE concepts should they use? (Choose three.)

Accepted Answer

Artifact types (e.g., Model, Metrics) for passing outputs. dsl.If for conditionals, Artifacts for passing model/data, and pre-built Vertex AI components for training.

Answer

Manual approval via Cloud Console

Answer

Cloud SQL for storing intermediate results