PDE Practice Test 18 — 15 Questions

Question 1

You need to orchestrate a simple, linear workflow that calls several Cloud Functions and API endpoints sequentially with conditional logic. The workflow should be defined as code and have minimal overhead. Which GCP service should you use?

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

Accepted Answer

Workflows. Workflows is a serverless orchestration service that uses YAML/JSON to define workflows. It is ideal for simpler, linear or conditional orchestrations without the need for full Airflow infrastructure.

Answer

Cloud Tasks

Answer

Dataflow

Answer

Cloud Composer

Question 2

A company has a trained model stored in Vertex AI Model Registry. They want to automate retraining when new training data arrives in Cloud Storage. Which approach is most efficient?

Accepted Answer

Use Cloud Functions triggered by Cloud Storage events to start a Vertex AI Training job. Cloud Functions can be directly triggered by Cloud Storage events (e.g., object finalize) to invoke the Vertex AI Training service via the AI Platform API. This creates an event-driven, serverless pipeline that retrains the model immediately when new data arrives, without polling or manual intervention, making it the most efficient and cost-effective approach.

Answer

Use Dataflow to continuously update the model

Answer

Use Cloud Scheduler to trigger a Cloud Build retraining step

Answer

Schedule a weekly Cloud Composer DAG to check for new data and retrain

Question 3

A data engineer notices that Spark jobs on the Dataproc cluster shown often fail with executor lost errors. What is the most likely reason?

Accepted Answer

All 10 workers are preemptible and can be reclaimed by Compute Engine at any time.. Preemptible VMs in Google Compute Engine can be terminated at any time due to resource contention or other factors, with only 30 seconds notice. If all 10 worker nodes are preemptible, Spark executors running on them will be frequently lost, causing job failures. This is the most direct cause of 'executor lost' errors in a Dataproc cluster.

Answer

The master node has only 4 vCPUs, which may be insufficient for job coordination.

Answer

The cluster is in a single zone, so a zone failure could cause all workers to shut down.

Answer

Autoscaling is enabled and scaling down is causing workers to be removed during job execution.

Question 4

Which BigQuery feature allows you to estimate the cost of a query before running it, by returning the number of bytes that would be processed?

Accepted Answer

--dry_run flag. The --dry_run flag in the BigQuery CLI or the dryRun parameter in the API simulates the query and returns the bytes processed without executing it, allowing cost estimation.

Answer

EXPLAIN statement

Answer

INFORMATION_SCHEMA.JOBS

Answer

Slot estimator

Question 5

You are building a Dataflow pipeline in Python that reads messages from Pub/Sub, enriches them with data from a BigQuery table, and writes the results to BigQuery. The enrichment lookup table is large and changes infrequently. Which approach minimizes cost and latency?

Accepted Answer

Use a side input that reads the BigQuery table periodically and caches it.. Using a side input that periodically reads the BigQuery table and caches it avoids querying BigQuery for every incoming message, which would be prohibitively expensive and high-latency. The side input is refreshed at a configurable interval (e.g., every 10 minutes) via a pipeline option, and the cached data is broadcast to all workers, enabling fast, in-memory lookups without per-element I/O. This approach minimizes cost by reducing BigQuery API calls and minimizes latency by avoiding synchronous queries for each message.

Answer

Use a CoGroupByKey transform to join the incoming stream with a stream from BigQuery.

Answer

Use BigQuery IO to query the table for every incoming message.

Answer

Use a stateful DoFn and store the lookup in state per key.

Question 6

A company runs a daily batch data processing pipeline using Cloud Dataproc. The pipeline reads 10 TB of CSV files from Cloud Storage, performs a heavy aggregation (GroupBy) and joins with a small reference table, then writes the results to BigQuery. The cluster consists of 20 n1-standard-8 nodes, including 10 preemptible workers for cost savings. Recently, the job completion time has doubled from 30 minutes to over an hour. The job logs show many tasks being retried, and the Shuffle spill ratio is high. No significant data volume change was observed. What is the most likely root cause?

Accepted Answer

Preemptible workers are being reclaimed, causing YARN container failures and task retries.. Preemptible workers are frequently reclaimed by Google Cloud, causing YARN containers to fail and tasks to be retried. This leads to increased job completion time and a high shuffle spill ratio, as partial shuffle data is lost and must be recomputed. The doubling of job time without data volume change strongly points to infrastructure instability rather than data or configuration issues.

Answer

The cluster's HDFS is running out of space due to intermediate shuffle data.

Answer

Data skew has developed, causing a few tasks to process most of the data.

Answer

The reference table has increased in size, causing more data to be broadcast to all workers.

Question 7

A company is designing a CI/CD pipeline for their ML models using Cloud Build and Vertex AI. Which TWO practices should they adopt to ensure reliable and reproducible deployments?

Accepted Answer

Use immutable container images with version tags for each model deployment. Using immutable container images with version tags ensures that each deployment is based on a fixed, unchangeable artifact. This eliminates the risk of configuration drift or unintended changes between builds, which is critical for reproducibility in Vertex AI deployments. Cloud Build can tag images with the commit SHA or build ID, making each deployment traceable and rollback-safe.

Answer

Require manual approval for every model change before deployment

Answer

Store all model artifacts in a single Cloud Storage bucket without versioning

Answer

Deploy every model version directly to production for immediate use

Question 8

A team runs a Dataflow streaming pipeline that reads from Pub/Sub, windows events by processing time, and writes to BigQuery. Some late-arriving events are being dropped. The requirement is to include all events that arrive within 10 minutes of the watermark. Which pipeline configuration should be used?

Accepted Answer

Use fixed windows with .withAllowedLateness(Duration.standardMinutes(10)). `withAllowedLateness(Duration.standardMinutes(10))` on a fixed window allows late-arriving events to be included up to 10 minutes after the watermark passes the window's end. This directly meets the requirement to retain events arriving within 10 minutes of the watermark, while still using processing-time windows as specified.

Answer

Use sliding windows with no allowed lateness

Answer

Use fixed windows with withAllowedLateness(Duration.standardSeconds(10))

Answer

Switch from processing time to event time and use default triggers

Question 9

A company is using Pub/Sub to ingest clickstream events. They need to ensure that events are delivered to a subscriber at least once, but duplicates can be tolerated. They also need to filter events by type before processing. Which subscription configuration should be used?

Accepted Answer

Pull subscription with a filter on event type attribute. Pub/Sub provides at-least-once delivery for both pull and push subscriptions. Filtering by attributes is supported at subscription level. Pull subscriptions are typically used when the subscriber controls the pace. Push subscriptions are also possible, but the question does not specify delivery method preference. The key is to enable message filtering on the subscription.

Answer

Pull subscription with exactly-once delivery enabled

Answer

Push subscription with no filter

Answer

Push subscription with a dead letter topic

Question 10

You are building a real-time fraud detection system using BigQuery streaming and a BQML logistic regression model. The model must be retrained every hour with new labeled data. What is the MOST cost-effective approach to serve predictions with low latency?

Accepted Answer

Export the model to a Cloud Storage bucket and deploy it to AI Platform Prediction. Exporting the model to Cloud Storage and deploying to AI Platform Prediction is the most cost-effective approach because AI Platform Prediction provides managed, autoscaling prediction serving with pay-per-prediction pricing. It avoids the cost and latency of repeatedly querying BigQuery with ML.PREDICT, which consumes slots and is not designed for real-time serving. Option C (Dataflow with model inference) incurs streaming pipeline costs, while options A and B are inefficient due to repeated BigQuery queries or unsupported materialized views with ML.PREDICT.

Answer

Call ML.PREDICT on a BigQuery table that is updated every hour

Answer

Use a BigQuery materialized view that refreshes every minute and apply ML.PREDICT

Answer

Stream data into Pub/Sub and use a Dataflow pipeline with Apache Beam's model inference

Question 11

A company wants to stream data from Cloud Pub/Sub into BigQuery with minimal latency. They have a small team and limited operational resources. Which approach is best?

Accepted Answer

Use a Dataflow pipeline with a BigQuery subscription.. A Dataflow pipeline with a BigQuery subscription provides a fully managed, serverless streaming solution that directly ingests messages from Pub/Sub and writes them to BigQuery with minimal latency. Dataflow handles autoscaling, checkpointing, and exactly-once semantics, which aligns with the team's limited operational resources. The BigQuery subscription (via the Pub/Sub to BigQuery template) eliminates the need for custom code or cluster management, ensuring low-latency streaming without operational overhead.

Answer

Write a custom application on Compute Engine that polls Pub/Sub and writes to BigQuery.

Answer

Create a Dataproc cluster running a Spark Streaming job.

Answer

Create a Cloud Function that writes to BigQuery.

Question 12

A data pipeline ingests streaming events into Pub/Sub. You need to guarantee that each event is processed exactly once downstream in Dataflow. Which combination of Pub/Sub and Dataflow configurations should you use?

Accepted Answer

Use Pub/Sub with a unique message ID and Dataflow with idempotent writes or Dataflow's exactly-once sink. Pub/Sub offers at-least-once delivery. To achieve exactly-once processing, the pipeline must be idempotent or use Dataflow's exactly-once sinks. Using a unique message ID for deduplication is a common approach.

Answer

Use Pub/Sub with exactly-once delivery enabled and Dataflow with exactly-once processing

Answer

Use Pub/Sub with message deduplication and Dataflow with at-least-once processing

Answer

Use Pub/Sub with a dead letter topic and Dataflow with automatic retries

Question 13

A company runs a batch processing job on Dataproc that uses Apache Spark to process 500 GB of data daily. The job completes successfully but takes 4 hours. The team wants to reduce the runtime to under 2 hours without increasing cost. What should they do?

Accepted Answer

Use preemptible VMs for worker nodes and increase the number of workers.. Preemptible VMs cost significantly less than standard VMs (about 60-80% discount). By using preemptible VMs for worker nodes, you can increase the number of workers (and thus parallelism) without increasing cost. This directly reduces runtime by distributing the 500 GB workload across more executors, while the cost savings from preemptible VMs offset the additional nodes.

Answer

Increase the master node's machine type to n2-standard-8.

Answer

Increase the machine type of worker nodes to n2-highmem-8.

Answer

Migrate the job to Dataflow with autoscaling enabled.

Question 14

A data engineer is using Vertex AI Workbench to develop a custom ML model. They want to store and version datasets, track experiments, and register models. Which three Vertex AI services should they use? (Choose THREE)

Accepted Answer

Vertex AI Model Registry. Vertex AI Dataset stores and manages datasets. Vertex AI Experiments tracks ML experiments. Vertex AI Model Registry stores and versions trained models.

Answer

Vertex AI Feature Store

Answer

Vertex AI Matching Engine

Question 15

A company runs a Dataflow streaming pipeline that reads from Pub/Sub and writes to BigQuery. They experience a sudden spike in data volume causing BigQuery write throughput to be exceeded, resulting in errors. Which strategy should they implement to handle this gracefully?

Accepted Answer

Use a BigQuery sink with 'WRITE_APPEND' mode and set 'writeDisposition' to 'WRITE_APPEND'.. Setting 'WRITE_APPEND' mode and 'writeDisposition' to 'WRITE_APPEND' configures the BigQuery sink to append data to the existing table. The default error handling in the BigQuery sink includes automatic retries and backpressure, which gracefully handle spikes in data volume. Options A, C, and D are incorrect because: A uses FAIL_FAST error handling, which fails the pipeline on write errors and does not route to a dead-letter queue; C truncates the table; D prevents table creation.

Answer

Use a BigQuery sink with 'FAIL_FAST' error handling and set a dead-letter queue for failed writes.

Answer

Use a BigQuery sink with 'WRITE_TRUNCATE' mode.

Answer

Use a BigQuery sink with 'CREATE_NEVER' write method.