PDE Practice Test 32 — 15 Questions

Question 1

A machine learning team wants to deploy a new model version for canary testing, where only 5% of traffic is routed to the new version. Which Vertex AI endpoint configuration supports this?

Accepted Answer

Configure the endpoint with traffic split: 95% to old version, 5% to new version.. Vertex AI endpoints natively support traffic splitting, allowing you to route a specified percentage of requests to different model versions deployed on the same endpoint. By configuring a traffic split of 95% to the old version and 5% to the new version, you can perform canary testing without additional infrastructure or client-side logic. This is the correct and simplest approach within Vertex AI.

Answer

Have the client application randomly select which model to call with 5% probability.

Answer

Deploy the new version to a separate endpoint and direct 5% of users via a load balancer.

Answer

Use an A/B testing framework outside of Vertex AI to compare results.

Question 2

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 3

A company has a BigQuery dataset containing sensitive customer data. They want to share a subset of this data with external partners, ensuring that partners can only see specific columns and rows. Which BigQuery feature should they use?

Accepted Answer

Authorized views. Authorized views in BigQuery allow a view to be created in one dataset that reads from a source dataset, and then access to the view is granted to external partners without granting access to the underlying source data. This enables column- and row-level filtering (via the view's SQL) so partners see only the permitted subset. This is the standard BigQuery pattern for sharing restricted data.

Answer

Materialized views

Answer

Clustered tables

Answer

Dataset-level access controls

Question 4

A company uses Cloud Spanner and needs to store a parent-child relationship where the child table is frequently queried together with the parent. The parent has millions of rows and the child billions. Which Spanner feature optimizes performance for this pattern?

Accepted Answer

Interleaved tables. Interleaved tables in Cloud Spanner physically co-locate child rows with their parent rows on the same split, so a parent-child join can be satisfied with a single local lookup rather than a distributed join across nodes. This is ideal when the child table is frequently queried together with the parent, as in this scenario with millions of parents and billions of children. The interleaving declaration (INTERLEAVE IN PARENT) makes the child's primary key prefix the parent's key, enabling efficient prefix scans and reduced network hops.

Answer

Partitioned tables

Answer

Secondary indexes

Answer

Change streams

Question 5

You are migrating on-premises Hadoop jobs to Google Cloud. The existing jobs use Spark for ETL and Hive for querying. You want to minimize changes to the existing code and maintain the ability to use Hive queries with the same metastore across multiple clusters. Which service combination should you use?

Accepted Answer

Cloud Dataproc with Cloud Storage and Dataproc Metastore. Cloud Dataproc runs managed Spark and Hive clusters, so existing Spark ETL jobs and Hive queries migrate with minimal code changes. Pairing Dataproc with Cloud Storage for data and Dataproc Metastore for a shared Hive metastore across clusters preserves the same table definitions and schema across multiple clusters.

Answer

Cloud Dataflow with Beam SQL

Answer

Cloud Dataproc with Dataproc on GKE

Answer

Cloud BigQuery with external tables on Cloud Storage

Question 6

You are designing a BigQuery data lake for a healthcare organization. The data includes patient records that must be access-controlled at the row level. Which TWO features should you use to meet this requirement?

Accepted Answer

Row-level security using row access policies. Row-level security using row access policies (A) is correct because BigQuery row access policies let you define a filter predicate (e.g., based on a patient_id or facility column) that restricts which rows each user or group can see, enforcing row-level access control directly on the table. Authorised views with row filters (B) are also correct because you can create a view that applies a WHERE clause to expose only permitted rows, then authorise that view on the dataset so users query the view without direct table access, achieving row-level control. Dataset-level IAM roles (C) only grant access at the dataset, table, or view level, not per row, so they cannot satisfy row-level requirements. Clustering on patient_id (D) is a physical storage optimization that improves query performance and reduces scanned bytes but provides no access control. Materialised views (E) cache precomputed query results for performance and do not enforce row-level security.

Answer

Dataset-level IAM roles

Answer

Clustering on patient_id

Answer

Materialised views

Question 7

Drag and drop the steps to configure a VPC network with private Google access for on-premises connectivity using Cloud VPN into the correct order.

Accepted Answer

Create VPC network with subnets, then enable Private Google Access on the subnets, then create Cloud VPN gateway and VPN tunnel, then configure the on-premises VPN device.. Private Google Access allows on-premises hosts to reach Google APIs via VPN without public IPs.

Answer

Create VPC network with subnets, then create Cloud VPN gateway and VPN tunnel, then enable Private Google Access on the subnets, then configure the on-premises VPN device.

Answer

Enable Private Google Access on the subnets, then create VPC network with subnets, then create Cloud VPN gateway and VPN tunnel, then configure the on-premises VPN device.

Answer

Create Cloud VPN gateway and VPN tunnel, then create VPC network with subnets, then enable Private Google Access on the subnets, then configure the on-premises VPN device.

Question 8

A data science team needs to ensure that a deployed Vertex AI model can handle varying traffic patterns with minimal latency and cost. What should they do?

Accepted Answer

Use Vertex AI Prediction with autoscaling. Vertex AI Prediction with autoscaling dynamically adjusts the number of serving instances based on incoming traffic, ensuring minimal latency during spikes and cost efficiency during lulls. This is the recommended approach for handling variable traffic patterns in production, as it leverages Google Cloud's managed infrastructure to scale from zero to thousands of nodes automatically.

Answer

Use batch prediction instead of online

Answer

Pre-warm all instances

Answer

Deploy to a single large machine type

Question 9

A company uses BigQuery ML to create a classification model. The model is used for batch prediction on a weekly basis. After six months, the data distribution shifts, and model accuracy drops. Which approach should the company take to maintain model performance?

Accepted Answer

Schedule automatic retraining of the model using the most recent three months of data.. The model's accuracy drop is due to data distribution shift (concept drift). Scheduling automatic retraining using the most recent three months of data ensures the model adapts to the new patterns without manual intervention. BigQuery ML supports scheduled queries and automatic model retraining via the `CREATE OR REPLACE MODEL` statement, making this approach both practical and aligned with MLOps best practices for batch prediction pipelines.

Answer

Use Cloud Dataflow to preprocess the data and then update the model with new features.

Answer

Perform hyperparameter tuning on the original training data.

Answer

Apply model quantization to reduce model size and improve inference speed.

Question 10

A production model deployed on Vertex AI Endpoint is experiencing high latency during traffic spikes. The current configuration uses a single replica. What is the most efficient solution?

Accepted Answer

Enable autoscaling with minReplicaCount=1 and maxReplicaCount=10. Enabling autoscaling with minReplicaCount=1 and maxReplicaCount=10 allows Vertex AI Endpoint to dynamically add replicas during traffic spikes, distributing inference requests across multiple instances and reducing latency. This is the most efficient solution as it scales resources up only when needed, avoiding over-provisioning and minimizing cost during low traffic periods.

Answer

Set a higher min replica count (e.g., 3)

Answer

Use a larger machine type (e.g., n1-highmem-8)

Answer

Switch to batch prediction to handle spikes

Question 11

A data engineer needs to enforce that all datasets in a project expire after 90 days to reduce storage costs. They want to automate this without manual intervention. Which approach should they use?

Accepted Answer

Set a default table expiration on each BigQuery dataset to 90 days. BigQuery datasets support a default table expiration setting that automatically deletes tables after a specified number of days. This enforces a 90-day lifecycle for all tables in the dataset without requiring manual intervention or external automation, directly addressing the cost reduction goal.

Answer

Use Cloud Storage lifecycle rules to delete tables after 90 days

Answer

Create an IAM policy that revokes access after 90 days

Answer

Use BigQuery scheduled queries to delete tables older than 90 days

Question 12

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 13

A Dataflow pipeline as described in the exhibit has increasing lag. Which optimization is most likely to reduce the lag?

Accepted Answer

Use FileLoads instead of StreamingInserts for BigQuery output. The exhibit shows increasing lag in a Dataflow pipeline writing to BigQuery. StreamingInserts (the default) use the BigQuery Storage Write API, which can throttle under high throughput, causing backpressure and lag. Switching to FileLoads writes data to temporary files in Cloud Storage and then loads them into BigQuery via batch load jobs, which decouples the write path from the streaming insert quota and reduces lag by avoiding per-row insert limits.

Answer

Increase the number of workers

Answer

Use global windows instead of fixed windows

Answer

Add additional ParDo transforms

Question 14

You have a BigQuery table that is partitioned by ingestion time and clustered on user_id. The table stores event logs and is queried frequently by user_id to analyze user behavior over the last 30 days. Queries are still scanning too many partitions. Which optimization should you apply first?

Accepted Answer

Change the partition column to a DATE column based on event_timestamp and keep clustering on user_id. Ingestion-time partitioning creates partitions based on when data arrives, not the event timestamp, so a query filtering on the last 30 days of events may still scan partitions that contain older events ingested recently. Changing the partition column to a DATE derived from event_timestamp aligns partition pruning with the query filter, and keeping clustering on user_id preserves the benefit for user-based filtering. This is the highest-impact first optimization.

Answer

Create a materialized view that pre-aggregates data by user_id and date

Answer

Remove partitioning and rely solely on clustering

Answer

Add clustering on a second column like event_type

Question 15

Which Google Cloud service would you use to create a unified data catalog that automatically captures lineage from BigQuery, Cloud Storage, and other sources?

Accepted Answer

Dataplex. Dataplex is Google Cloud's unified data governance and catalog service that automatically discovers, catalogs, and captures lineage across BigQuery, Cloud Storage, and other sources. It provides a unified data catalog with automatic metadata extraction and lineage tracking, which is exactly what the question describes.

Answer

Cloud Composer

Answer

Dataflow

Answer

Data Catalog