PDE Practice Test 43 — 15 Questions

Question 1

A data engineer needs to schedule a nightly transfer of data from an Amazon S3 bucket to Cloud Storage. Which two steps are required to achieve this? (Choose TWO.)

Accepted Answer

Grant appropriate permissions to the transfer service account. The Storage Transfer Service uses a Google-managed service account to access the source S3 bucket. You must grant this service account the appropriate IAM permissions (e.g., `s3:GetObject` and `s3:ListBucket`) on the S3 bucket via an AWS IAM policy. Without these permissions, the transfer job cannot read the data from S3.

Answer

Configure a VPC peering between AWS and GCP

Answer

Use gsutil rsync in a cron job

Answer

Set up a Cloud Function to copy files

Question 2

You have deployed a classification model on Vertex AI Endpoints. The model's training data had a balanced class distribution, but over time, the production data has shifted such that one class appears 90% of the time. The model's overall accuracy remains high, but the recall for the minority class has dropped significantly. What is the best approach to detect and address this issue?

Accepted Answer

Set up Vertex AI Model Monitoring to detect skew and drift, and retrain using a sliding window of recent data. Vertex AI Model Monitoring is specifically designed to detect skew and drift between training and serving data. In this scenario, the production data has shifted to 90% of one class, which is a clear case of data drift. By setting up monitoring, you can be alerted to this drift and then retrain the model using a sliding window of recent data, which adapts to the new distribution without requiring full retraining on the entire historical dataset. This approach directly addresses the root cause—the shift in class distribution—rather than just treating symptoms.

Answer

Retrain the model daily on the entire historical dataset

Answer

Increase the number of replicas on the endpoint to reduce latency

Answer

Adjust the decision threshold to improve minority class recall

Question 3

You manage a BigQuery reservation with 500 baseline slots and autoscaling up to 2000 slots. Your team runs a mix of interactive queries and batch load jobs. During peak hours, you notice that interactive queries are throttled when autoscaling slots are consumed by long-running batch loads. How can you ensure interactive queries get priority access to slots?

Accepted Answer

Create a separate reservation for interactive queries with a higher priority assignment.. BigQuery reservations allow you to create separate reservations for different workloads (e.g., interactive queries vs. batch loads) and assign them different priority levels. By creating a dedicated reservation for interactive queries with a higher priority, you ensure that interactive queries get preferential access to slots, even when autoscaling slots are consumed by long-running batch jobs. This directly addresses the contention issue without reducing overall capacity.

Answer

Reduce the baseline slots to 200 and rely solely on autoscaling.

Answer

Switch to on-demand pricing to eliminate slot contention.

Answer

Set the autoscaling max to 1000 slots for batch jobs.

Question 4

A team is deploying a TensorFlow model for online predictions on AI Platform Prediction. They want to monitor for data drift and model performance degradation. Which TWO Google Cloud services should they use?

Accepted Answer

AI Platform Continuous Evaluation. AI Platform Continuous Evaluation (option B) is correct because it is a managed service specifically designed to detect data drift and model performance degradation in deployed models. It automatically compares incoming prediction data against the training data distribution and monitors metrics like accuracy over time, triggering alerts when significant drift is detected. Cloud Monitoring (option C) is correct because it provides the underlying metrics and alerting infrastructure that can track model performance indicators (e.g., prediction latency, error rates) and integrate with Continuous Evaluation for comprehensive observability.

Answer

Cloud Composer

Answer

AI Platform Pipelines

Answer

Cloud Logging

Question 5

A company has a Cloud Functions function that triggers on new files in Cloud Storage and writes a message to Pub/Sub for downstream processing. Recently, the function has been timing out after 60 seconds. The downstream processing is critical. What is the best solution?

Accepted Answer

Increase function timeout to 540 seconds and delegate heavy processing to Cloud Dataflow. Cloud Functions has a maximum timeout of 540 seconds (9 minutes) for HTTP-triggered functions, and by increasing the timeout you allow the function to complete its work. Delegating heavy processing to Cloud Dataflow offloads the computationally intensive tasks, preventing future timeouts and ensuring scalable, reliable downstream processing for critical workloads.

Answer

Replace Cloud Functions with a Cloud Run job that has longer timeout

Answer

Increase the function memory to 2 GB to speed up execution

Answer

Reduce the function timeout to 30 seconds to force faster execution

Question 6

A company uses GKE to run microservices. They want to ensure the application restarts automatically if it becomes unresponsive. Which probes should they configure in their pod spec?

Accepted Answer

Both readiness and liveness probes.. To ensure a pod restarts automatically when it becomes unresponsive, you need a liveness probe. However, the question asks which probes to configure, and the best practice is to use both readiness and liveness probes. The liveness probe restarts the container if it fails, while the readiness probe controls traffic routing, preventing requests from being sent to an unresponsive pod before the liveness probe triggers a restart. Option D is correct because it combines both for robust health management.

Answer

Startup probes only.

Answer

Liveness probes only.

Answer

Readiness probes only.

Question 7

You run batch predictions using Vertex AI Batch Prediction on a tabular dataset. The job processes 1 million rows and takes 6 hours to complete. You need to reduce the processing time to under 2 hours without increasing cost significantly. What should you do?

Accepted Answer

Increase the machine count (number of worker replicas) in the batch prediction job. Vertex AI Batch Prediction supports distributed processing by increasing the number of worker replicas. By adding more workers, the job can process partitions of the 1 million rows in parallel, reducing wall-clock time from 6 hours to under 2 hours. This approach scales horizontally without requiring more expensive machine types, keeping costs roughly linear with the number of workers and avoiding the diminishing returns of vertical scaling.

Answer

Switch to a machine type with more CPU cores and vCPUs

Answer

Downsample the dataset to 500k rows

Answer

Use Online prediction instead of batch

Question 8

Your streaming Dataflow pipeline reads from Pub/Sub, enriches data with a side input, and writes to BigQuery. You need to update the enrichment logic without draining the pipeline, to minimize data loss and maintain exactly-once semantics. What should you do?

Accepted Answer

Use the Dataflow job update mechanism to replace the pipeline with a new version.. The Dataflow job update mechanism allows you to replace a running pipeline's code with a new version without draining or stopping it, preserving the existing state and minimizing data loss. This mechanism supports exactly-once semantics by ensuring that all in-flight elements are processed exactly once, even after the update, by maintaining the pipeline's checkpoint and watermark state.

Answer

Cancel the pipeline and create a new one with the updated code.

Answer

Stop the pipeline, update the code, and restart from the latest snapshot.

Answer

Drain the pipeline, update the code, and restart with the same job ID.

Question 9

A company is designing a data processing system that must handle both batch and streaming workloads with unified pipeline code. Which two Google Cloud services are most suitable for implementing a unified batch and streaming pipeline? (Choose TWO.)

Accepted Answer

Apache Beam SDK. Apache Beam SDK (C) provides a unified programming model that allows developers to write a single pipeline that can execute in both batch and streaming modes without code changes. It abstracts the underlying execution engine, making it the correct choice for unified pipeline code.

Answer

Cloud Data Fusion

Answer

BigQuery

Answer

Cloud Dataproc

Question 10

An organization needs to store transactional data for a global e-commerce platform with strong consistency across regions and an SLA of 99.999% availability. The application requires SQL semantics with horizontal scaling. Which Google Cloud database should they choose?

Accepted Answer

Cloud Spanner. Cloud Spanner is the correct choice because it provides globally distributed, strongly consistent SQL semantics with horizontal scaling and a 99.999% availability SLA. It uses synchronous replication and the TrueTime API to ensure external consistency across regions, meeting the strict consistency and uptime requirements of a global e-commerce platform.

Answer

Firestore

Answer

Cloud SQL

Answer

Cloud Bigtable

Question 11

A company uses Dataproc for transient clusters. Which TWO actions can reduce costs?

Accepted Answer

Set cluster autoscaling to minimize idle resources. Dataproc cluster autoscaling automatically adjusts the number of worker nodes based on the YARN memory and CPU utilization metrics. By scaling down during idle periods, you avoid paying for unused compute capacity, directly reducing costs for transient clusters that have variable workloads.

Answer

Increase master node size

Answer

Use standard VMs for all nodes

Answer

Use persistent clusters to avoid creation overhead

Question 12

You are building a time-series forecasting model with BigQuery ML. Which three steps should you perform to properly split the data and evaluate the model? (Choose THREE)

Accepted Answer

Evaluate on a holdout set that is later in time than the training set.. For time-series, you must maintain temporal order: split chronologically (not randomly), use a cutoff date for training/validation, and evaluate on unseen future data. Cross-validation should be time-series aware (e.g., expanding window). Random split is invalid for time-series. Using a single train/test split may be insufficient; multiple windows are better.

Answer

Split the data randomly into training and testing sets.

Answer

Use k-fold cross-validation with random folds.

Question 13

A data engineer is designing a pipeline that reads from Cloud Pub/Sub, aggregates events into 5-minute windows, and writes the results to BigQuery. The engineer wants to ensure that late-arriving data (up to 2 minutes late) is included in the correct window. Which Dataflow feature should they configure?

Accepted Answer

Set the allowed lateness to 2 minutes with a trigger that fires on late data. Dataflow's allowed lateness feature (set to 2 minutes) ensures that late-arriving data within that threshold is still assigned to the correct 5-minute window. Combined with a trigger that fires on late data, the pipeline can emit updated results for the window after the watermark passes, which is exactly what the engineer needs to handle late-arriving events up to 2 minutes late.

Answer

Use a sliding window of 5 minutes with 2-minute slide

Answer

Set the window duration to 7 minutes to account for lateness

Answer

Use a global window and watermark

Question 14

A data engineer needs to transfer 500 TB of archival data from an on-premises NAS to Cloud Storage. The on-premises network has limited bandwidth (100 Mbps). Which transfer method should they recommend?

Accepted Answer

Transfer Appliance. The Transfer Appliance is a physical device designed for large-scale data transfers (up to petabytes) when network bandwidth is insufficient. With 500 TB of data and only 100 Mbps bandwidth, the theoretical transfer time would be over 500 days, making any online transfer method impractical. The Transfer Appliance bypasses network constraints entirely by shipping the data physically to Google Cloud.

Answer

Storage Transfer Service for on-premises

Answer

gsutil rsync

Answer

Dataflow pipeline reading from NAS

Question 15

A financial services company must comply with GDPR "right to be forgotten". They store customer transactions in BigQuery partitioned by date. When a user requests deletion, all their data must be removed within 48 hours. The deletion requests are received via a Pub/Sub topic. What is the most scalable and cost-effective approach?

Accepted Answer

Use a Dataflow pipeline that reads the deletion IDs from Pub/Sub, joins with the transactions table using a side input, and writes the filtered data to a new table, then swapping. It uses Dataflow to process deletion requests from Pub/Sub, join them with the BigQuery transactions table via a side input, and write a filtered copy to a new table. This approach is scalable (handles high-throughput streaming deletions) and cost-effective (avoids expensive DELETE mutations on BigQuery, which consume slot resources and can be slow for large tables). Swapping the new table for the old one completes the deletion efficiently within the 48-hour SLA.

Answer

Use Cloud Functions to execute a BigQuery DELETE statement on each request

Answer

Use Cloud DLP to redact the user's data in Cloud Storage

Answer

Use BigQuery table snapshots and restore after deletion