PDE Practice Test 4 — 25 Questions

Question 1

A startup is building a mobile app that needs to sync user data across devices in real time. They expect millions of concurrent users and need a NoSQL database with offline support and automatic multi-region replication. Which Google Cloud service meets these requirements?

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

Accepted Answer

Firestore. Firestore is a NoSQL, serverless document database that provides real-time synchronization, offline support via local persistence, and automatic multi-region replication. It is designed for mobile and web apps with millions of concurrent users, making it the ideal choice for this use case.

Answer

Cloud Bigtable

Answer

Cloud Spanner

Answer

Cloud SQL

Question 2

A data scientist needs to perform feature engineering for a machine learning model using Vertex AI. They want to preprocess data using a pipeline that includes scaling, one-hot encoding, and handling missing values. Which TWO services can they use to define and execute this preprocessing pipeline? (Choose 2.)

Accepted Answer

Vertex AI Pipelines. Vertex AI Pipelines is the recommended service for building and running ML pipelines, including preprocessing steps. Alternatively, you can use BigQuery SQL for feature engineering directly on the data, then export the processed data for training. Cloud Dataflow is an option for batch/streaming data processing but is not specific to ML pipelines. Cloud Functions and Dataproc are less suitable for this purpose.

Answer

Cloud Dataproc

Answer

Cloud Dataflow

Answer

Cloud Functions

Question 3

A data analyst needs to transform nested and repeated fields in BigQuery. They have a table with a column of type ARRAY>. Which SQL function should they use to flatten the array into individual rows for analysis?

Accepted Answer

UNNEST. UNNEST is used to flatten arrays into rows. STRUCT is used to group fields. CAST is for type conversion. REPLACE is for string replacement.

Answer

STRUCT

Answer

CAST

Answer

REPLACE

Question 4

A data engineer needs to create a BigQuery table that is partitioned by ingestion time and clustered by customer_id and transaction_date. They also want to limit access so that only users from a specific domain can query the table. Which approach should they use?

Accepted Answer

Create the table with partitioning and clustering, then create an authorized view on the table and grant the view access to the domain users. Authorized views allow sharing query results with specific users/groups without giving direct table access. Clustering and partitioning are defined at table creation. IAM roles at dataset level are too broad. Row-level security filters rows but doesn't restrict domain.

Answer

Create the table with partitioning only, then use a materialized view to restrict access

Answer

Create the table without clustering, use row-level security to filter by domain, and grant access to the table

Answer

Create the table with partitioning and clustering, then grant bigquery.dataViewer to the domain via IAM at the dataset level

Question 5

A data engineering team is operationalizing a machine learning model for real-time inference. They need to monitor the model's performance in production. Which THREE types of monitoring should they implement? (Choose three.)

Accepted Answer

Model accuracy decay. Model accuracy decay (A) is critical because in production, the model's predictive performance can degrade over time due to changes in the underlying data distribution or business logic. Monitoring accuracy decay allows the team to detect when the model no longer meets its performance baseline, triggering retraining or rollback. This is a standard practice in MLOps for maintaining model reliability.

Answer

Model re-training frequency

Answer

Training pipeline failures

Question 6

A company is building a real-time streaming pipeline using Pub/Sub and Dataflow to process clickstream data. The pipeline writes aggregated metrics to BigQuery every 10 seconds using a fixed window. During peak traffic, some windows produce duplicate rows in BigQuery. What is the most likely cause?

Accepted Answer

Dataflow is retrying BigQuery streaming inserts after a timeout, and the retries succeed even though the original insert succeeded.. Dataflow uses at-least-once semantics for streaming inserts into BigQuery. When a streaming insert times out, Dataflow retries the insert, and if the original insert actually succeeded but the acknowledgment was lost, the retry produces a duplicate row. This is a known behavior of BigQuery streaming inserts with retry logic.

Answer

The pipeline uses default triggers instead of after-watermark triggers.

Answer

The fixed window duration is too short, causing overlapping windows.

Answer

The pipeline is using too many Dataflow workers, causing load balancing issues.

Question 7

A data pipeline is built with Cloud Dataflow that reads from Pub/Sub, applies transformations, and writes to BigQuery. The pipeline is experiencing high latency and occasional data loss during worker failures. The engineer wants to improve reliability and performance. Which two actions should they take?

Accepted Answer

Enable Dataflow Streaming Engine and use BigQuery exactly-once sink. Enabling streaming engine moves the state management to the backend, reducing latency and improving reliability. Using exactly-once sinks (like BigQuery with exactly-once guarantees) prevents data loss.

Answer

Switch to Cloud Dataproc and use Spark Structured Streaming

Answer

Increase the number of workers and use at-least-once delivery to BigQuery

Answer

Use a global window and disable triggers

Question 8

A company wants to monitor the performance of a deployed model in production. Which metric indicates that the model's predictions are degrading?

Accepted Answer

Increase in prediction error rate. An increase in prediction error rate directly indicates that the model's outputs are deviating from the expected or ground-truth values, signaling degradation in predictive performance. This metric captures the core concept of model drift, where the statistical properties of the input data or the relationship between features and labels change over time, leading to less accurate predictions. In production ML monitoring, tracking error rate (e.g., classification accuracy, RMSE) is the primary method to detect when a model needs retraining or updating.

Answer

Increase in prediction latency

Answer

Decrease in throughput

Answer

Increase in number of requests

Question 9

You are a machine learning engineer at a FinTech company. Your team has developed a credit risk model using XGBoost and deployed it on Vertex AI Prediction using a custom container. The model is used for real-time credit decisions, and the endpoint is configured with a single machine type (n1-standard-4) and min_replica_count = 2, max_replica_count = 10. Recently, the team observed that during a promotional campaign, the endpoint's prediction latency increased from 200ms to over 2 seconds, and some requests resulted in 503 errors. You check the Cloud Monitoring metrics and see that CPU utilization reached 100% on the existing replicas, but the number of replicas never scaled beyond the initial 2. The deployment uses a custom container that runs a TensorFlow Serving-like model server. The container image is stored in Artifact Registry. The Vertex AI endpoint is configured with a traffic split of 100% to this model version. What is the most likely cause of the scaling failure, and what step should you take to resolve it?

Accepted Answer

Ensure the custom container exposes the correct metrics for CPU utilization so that Vertex AI autoscaling can trigger.. Vertex AI's autoscaling relies on the custom container exposing standard metrics (e.g., CPU utilization via the /metrics endpoint in a Prometheus format or through the Vertex AI custom metric adapter). If the container does not expose these metrics, the autoscaler cannot detect high CPU usage and will not trigger scaling beyond the initial replicas, leading to latency spikes and 503 errors under load.

Answer

Increase min_replica_count to 5 to handle the baseline load.

Answer

Change the endpoint configuration to use gRPC instead of HTTP to reduce latency.

Answer

Set the max_replica_count to a higher value like 20.

Question 10

You need to process a large volume of event data from Cloud Storage, apply complex transformations using Apache Spark, and then load the results into BigQuery. The data arrives in batches every hour. You want to minimize costs by using preemptible VMs. Which service should you use?

Accepted Answer

Dataproc. Dataproc supports preemptible (now called spot) VMs for cost savings. Dataflow does not support preemptible VMs for workers; it uses standard VMs. Cloud Composer is orchestration only. BigQuery is not for running Spark.

Answer

Cloud Composer

Answer

BigQuery

Answer

Dataflow

Question 11

A company is running a Cloud Dataflow streaming pipeline that aggregates events in 1-minute windows. They notice that the watermark is lagging significantly behind real-time. What is the most likely cause?

Accepted Answer

A hot key is causing data skew.. A hot key causes data skew, which means a disproportionate amount of data is assigned to a single key. In Cloud Dataflow, this leads to a single worker processing the bulk of the events, creating a processing bottleneck. The watermark, which tracks the progress of event-time processing, cannot advance until all data for a given window is processed, so the skewed key delays watermark progression significantly behind real-time.

Answer

The window duration is too short.

Answer

The pipeline was recently updated.

Answer

The allowed lateness is set too high.

Question 12

You are designing a near-real-time CDC pipeline to replicate changes from an on-premises PostgreSQL database to BigQuery for analytics. The source database has high transaction volume and you must ensure minimal impact on the source. Which Google Cloud service should you use to ingest the change data?

Accepted Answer

Datastream to stream changes to GCS, then load into BigQuery.. Datastream is purpose-built for CDC from MySQL, PostgreSQL, and Oracle to BigQuery or GCS. It reads the database logs (e.g., WAL) to capture changes with low latency and minimal impact on the source.

Answer

Pub/Sub with a custom connector that polls the database every minute.

Answer

Use BigQuery Data Transfer Service for PostgreSQL.

Answer

Use Dataflow with a JDBC IO connector to read from PostgreSQL.

Question 13

A data engineer is designing a streaming pipeline using Dataflow with Apache Beam. The pipeline reads from Pub/Sub, performs a stateful transformation (e.g., session windowing), and writes to BigQuery. The pipeline must handle late data and ensure exactly-once semantics. Which THREE configurations are required?

Accepted Answer

Set allowed lateness on the window to accommodate late data. Exactly-once sink (BigQuery) ensures no duplicates. Idempotent writes are built into Dataflow's BigQuery sink when using exactly-once mode. Setting allowed lateness handles late data. Watermark estimation is automatic; custom is not required. Triggers are optional; default trigger works. Windowing is inherent.

Answer

Use the File Loads write method for BigQuery

Answer

Use a custom watermark estimation function for Pub/Sub source

Question 14

A company runs a real-time anomaly detection system on Google Cloud. Streaming data from IoT devices is ingested via Pub/Sub, processed by Dataflow (Apache Beam), and results are written to Bigtable for low-latency serving. Recently, the system has been experiencing increased latency and occasional data loss. The Dataflow pipeline shows high system lag and backlog in Pub/Sub. The Bigtable cluster has 3 nodes and is reporting high CPU utilization (over 90%). The team suspects the issue is with the pipeline configuration. They have already verified that there are no errors in the pipeline code and no network issues. Which action should they take to resolve the issue?

Accepted Answer

Increase the number of Bigtable nodes to handle the write throughput.. The high CPU utilization on Bigtable (over 90%) indicates that the cluster is saturated and cannot keep up with the write throughput from Dataflow. This causes backpressure in the pipeline, leading to increased system lag and backlog in Pub/Sub, and eventually data loss when Pub/Sub messages expire. Increasing the number of Bigtable nodes directly addresses the bottleneck by distributing the write load and reducing CPU pressure, which allows the pipeline to drain the backlog and reduce latency.

Answer

Change the Dataflow worker machine type to n2-standard-8.

Answer

Decrease the batch size in the Dataflow pipeline to reduce latency.

Answer

Increase the number of Dataflow workers to process messages faster.

Question 15

A company wants to transfer 500 TB of data from an on-premises Hadoop cluster to Google Cloud Storage (GCS) for processing with Dataproc. The on-premises network has a 1 Gbps dedicated link to Google Cloud. The data must be transferred as quickly as possible, minimizing network usage. Which transfer method should they use?

Accepted Answer

Use Transfer Appliance to physically ship the data.. Transfer Appliance is the correct method because the dataset is 500 TB and the network link is only 1 Gbps. At 1 Gbps, the theoretical maximum transfer time is over 46 days, and real-world throughput (due to overhead, congestion, and Hadoop data characteristics) would be even longer. Transfer Appliance physically ships the data, bypassing the network bottleneck entirely and minimizing network usage, which is the stated requirement.

Answer

Use Storage Transfer Service over the 1 Gbps link.

Answer

Use gsutil cp in parallel with multiple threads.

Answer

Use BigQuery Data Transfer Service for Hadoop.

Question 16

A company is building a data lake on Cloud Storage with data from multiple sources. They need to apply schema-on-read and support ad-hoc SQL queries. Which architecture is most suitable?

Accepted Answer

Ingest to Cloud Storage, create BigQuery external tables.. BigQuery external tables allow schema-on-read by defining the schema at query time over data stored in Cloud Storage, enabling ad-hoc SQL queries without loading data into a separate system. This architecture directly supports the requirement for schema-on-read and SQL-based analysis, as BigQuery provides a serverless, scalable SQL engine.

Answer

Ingest to Cloud Spanner, query directly.

Answer

Ingest to Cloud SQL, then export to Cloud Storage for queries.

Answer

Ingest to Cloud Storage, load into Dataproc for queries.

Question 17

Which TWO factors should be considered when choosing between Cloud Dataflow and Dataproc for a batch processing pipeline?

Accepted Answer

Dataflow is built for data processing patterns, while Dataproc is better for general-purpose compute.. Dataflow is purpose-built for data processing patterns like batch and stream processing with unified programming models (Apache Beam), while Dataproc is optimized for general-purpose compute workloads such as running custom Spark, Hadoop, or ML jobs. Option D is correct because Dataflow provides automatic horizontal autoscaling based on pipeline throughput, whereas Dataproc requires manual cluster sizing or configuration of autoscaling policies, which are not as granular or reactive as Dataflow's.

Answer

Dataproc allows custom Docker containers, while Dataflow does not.

Answer

Dataproc supports Python, while Dataflow only supports Java.

Answer

Dataflow supports Java and Python, while Dataproc only supports Java.

Question 18

You have deployed a TensorFlow model on Vertex AI Endpoints with autoscaling. The model receives high traffic during peak hours, but you notice that inference latency increases significantly during cold starts. Which strategy would best minimize cold-start latency without incurring unnecessary cost?

Accepted Answer

Set minNodes to a value that handles baseline traffic, and use traffic splitting to gradually shift traffic to new replicas. Setting minNodes to a value that handles baseline traffic ensures that a minimum number of replicas are always warm, eliminating cold starts for baseline requests. Traffic splitting gradually shifts new traffic to newly created replicas, allowing them to warm up before receiving full load, which minimizes latency spikes without over-provisioning resources.

Answer

Set minNodes to 0 and enable node auto-scaling

Answer

Increase maxNodes to allow more replicas during peak, and rely on Kubernetes Horizontal Pod Autoscaler

Answer

Use Cloud Functions with Cloud Run for the model inference to leverage serverless cold-start mitigation

Question 19

A company is migrating ML workflows to Vertex AI Pipelines. They want to ensure best practices for pipeline reproducibility and debugging. Which THREE actions should they take? (Choose three.)

Accepted Answer

Set a random seed for all training components. Setting a random seed for all training components ensures deterministic behavior, meaning that the same inputs will produce the same outputs across multiple runs. This is critical for debugging and reproducibility in Vertex AI Pipelines, as it eliminates stochastic variability that can mask bugs or make results irreproducible. Without a fixed seed, even identical code and data can yield different model weights or metrics, complicating root cause analysis.

Answer

Use dynamic pipeline parameters for each run

Answer

Use conditional execution based on previous component outputs

Question 20

A data pipeline uses Cloud Composer to orchestrate Dataflow and BigQuery jobs. The pipeline fails intermittently with dependency errors. Which design change can improve reliability?

Accepted Answer

Use retries with exponential backoff. Cloud Composer (Apache Airflow) tasks can fail due to transient issues like API rate limits or resource contention. Implementing retries with exponential backoff allows the DAG to automatically re-attempt failed tasks with increasing delays, reducing the impact of intermittent failures without manual intervention. This is a standard Airflow pattern for improving reliability in orchestrated pipelines.

Answer

Switch to Cloud Functions for orchestration

Answer

Increase worker count in Dataflow

Answer

Use a simpler DAG with fewer dependencies