GCP-ADP Data Preparation And Ingestion Practice Test 4 — 15 Questions

Question 1

You need to ingest IoT data into BigQuery. You need to handle messages arriving in millions per second. Which service acts as the buffer?

Accepted Answer

Pub/Sub. Pub/Sub is Google Cloud's fully managed, globally scalable messaging service designed to decouple producers from consumers, and it can absorb millions of messages per second without backpressure. In an IoT ingestion architecture, Pub/Sub acts as the durable buffer that holds incoming telemetry until BigQuery (via Dataflow or the Storage Write API) can commit it. This decoupling prevents data loss during BigQuery load spikes and lets each side scale independently.

Answer

Cloud SQL

Answer

Cloud Functions

Answer

BigQuery Streaming API

Question 2

You are running a Dataflow job and notice that it is consuming too much memory and crashing. How do you optimize this?

Accepted Answer

Select a higher memory machine type. Dataflow worker memory is determined by the machine type; if a job is OOMing, selecting a higher-memory machine type (e.g., n1-highmem or a custom machine type) gives each worker more RAM to hold state, buffers, and user code. This directly addresses the memory pressure without changing parallelism.

Answer

Increase the number of workers

Answer

Reduce the number of workers

Answer

Use Batch mode

Question 3

You need to track who accessed which GCS bucket. Which service provides this audit trail?

Accepted Answer

Cloud Audit Logs. Cloud Audit Logs is the correct answer because it automatically records all administrative and data access activities on GCS buckets, including who accessed what, when, and from where. Specifically, Admin Activity logs capture bucket configuration changes, while Data Access logs (when enabled) record object reads and writes. This provides the audit trail required to track access to GCS buckets.

Answer

Cloud Security Command Center

Answer

Cloud Monitoring

Answer

Cloud Trace

Question 4

You are processing streaming data in Pub/Sub and need to archive every message into GCS without writing custom code. What should you use?

Accepted Answer

Pub/Sub Cloud Storage subscription. A Pub/Sub Cloud Storage subscription is a built-in subscription type that writes messages directly to a GCS bucket without any custom code. You configure the subscription with a bucket, filename prefix/suffix, and format, and Pub/Sub handles batching and delivery. This satisfies the requirement of archiving every message to GCS with no custom code.

Answer

Dataflow

Answer

BigQuery Data Transfer Service

Answer

Cloud Functions

Question 5

You need to ensure that deleted files in GCS can be restored for 30 days. What should you configure?

Accepted Answer

Object Versioning. Object Versioning preserves prior versions of objects when they are overwritten or deleted, allowing restoration for as long as the noncurrent versions are retained. To meet the 30-day restore requirement, you enable Object Versioning on the bucket and optionally add a lifecycle rule to delete noncurrent versions after 30 days. This directly satisfies the recoverability requirement without locking the bucket.

Answer

Cloud Storage bucket lock

Answer

Bucket Lifecycle policy

Answer

IAM

Question 6

You are cleaning data using Dataprep. You want to save the final dataset in BigQuery. What do you do?

Accepted Answer

Use the Export feature. In Dataprep (now Dataprep by Trifacta), after cleaning and transforming a dataset in the flow, you publish results using the Export (or 'Run job'/output) feature, where you specify BigQuery as the output destination and configure the target dataset and table. This writes the transformed data into BigQuery. Copying files or running SQL manually is not how Dataprep delivers results.

Answer

Copy the file from GCS

Answer

Use a SQL query

Answer

Wait for Dataflow to finish

Question 7

You are using Dataflow to read from Pub/Sub. The pipeline is failing due to malformed messages. How can you handle these without crashing?

Accepted Answer

Use a Dead Letter Queue (DLQ). A Dead Letter Queue pattern in Dataflow captures messages that fail transformation or parsing so the pipeline can continue processing valid records. In Apache Beam you implement this by branching the pipeline: successful records go to the main sink, and failed records (caught via try/except or a tagged output) are written to a separate Pub/Sub topic or BigQuery table for later inspection. This prevents a single malformed message from killing the streaming job.

Answer

Ignore errors

Answer

Increase memory

Answer

Restart the pipeline

Question 8

What is the primary benefit of using partitioned tables in BigQuery?

Accepted Answer

Performance and cost reduction. Partitioned tables in BigQuery split data into segments based on a column (e.g., date, timestamp, integer range), so queries with filters on the partition column scan only relevant partitions. This dramatically reduces bytes processed, improving query performance and lowering on-demand costs, which are billed per byte scanned.

Answer

Data security

Answer

Increased storage capacity

Answer

Data integrity

Question 9

You need to ingest log data into BigQuery from GCS files that arrive sporadically. Which service is best to trigger this?

Accepted Answer

Cloud Functions. Cloud Functions is an event-driven serverless compute service that can be triggered by changes in Cloud Storage, such as when a new file arrives. It is ideal for sporadic, unpredictable file arrivals because it scales automatically and you only pay for execution time. You can write a function that triggers on the object finalize event and loads the data into BigQuery.

Answer

Cloud Scheduler

Answer

BigQuery Data Transfer Service

Answer

Dataflow

Question 10

You are running a Dataflow job that joins two large datasets. Which join strategy should you avoid to prevent OOM errors?

Accepted Answer

Side Input Join with a large dataset. A Side Input Join involves loading one dataset as a side input into memory, which is then available to each worker. If the side input dataset is large, it can cause out-of-memory (OOM) errors on workers because the entire dataset must fit in memory. Therefore, using a side input join with a large dataset is the strategy to avoid. Other options like CoGroupByKey are designed for large-scale joins.

Answer

Flatten

Answer

CoGroupByKey

Answer

ParDo

Question 11

Which GCP tool provides a managed way to run Apache Airflow workflows?

Accepted Answer

Cloud Composer. Cloud Composer is a fully managed workflow orchestration service built on Apache Airflow. It handles the provisioning, scaling, and maintenance of Airflow infrastructure, allowing you to author, schedule, and monitor workflows (DAGs) using Python. It integrates natively with other GCP services like BigQuery, Dataflow, and Cloud Storage, making it the go-to choice for orchestrating complex data pipelines.

Answer

Dataflow

Answer

Cloud Scheduler

Answer

Dataprep

Question 12

You need to anonymize PII (Personally Identifiable Information) before loading data into BigQuery. Which tool is best?

Accepted Answer

Cloud DLP. Cloud DLP (Data Loss Prevention) is purpose-built for discovering, classifying, and de-identifying sensitive data such as PII. It provides de-identification techniques like masking, tokenization, and format-preserving encryption that can be applied to data before it lands in BigQuery. Unlike general-purpose data processing tools, Cloud DLP natively understands PII types (e.g., names, credit card numbers, SSNs) and offers configurable infoType detectors, making it the most efficient and compliant choice for anonymization.

Answer

Dataflow

Answer

BigQuery

Answer

Cloud Storage

Question 13

You need to delete all files in a GCS bucket older than 90 days. What is the most efficient approach?

Accepted Answer

Cloud Lifecycle Management. Cloud Lifecycle Management is the most efficient approach because it lets you define a lifecycle rule on the bucket that automatically deletes objects older than 90 days, with no code, no compute, and no manual intervention. It is a native GCS feature designed exactly for age-based object deletion, applying at scale across all objects in the bucket.

Answer

Manual deletion in the console

Answer

Use Dataflow to filter files

Answer

Write a cron job with gsutil

Question 14

You are using Dataflow with autoscaling. What metric determines whether the worker pool scales up or down?

Accepted Answer

Work backlog and throughput. Dataflow autoscaling is driven by the work backlog (unprocessed elements) and throughput relative to the current worker pool. The autoscaler compares the backlog and the rate of processing to decide whether to add or remove workers, targeting a steady-state where backlog stays low without over-provisioning. This is the documented signal for Streaming Engine and batch autoscaling.

Answer

Network latency

Answer

Memory usage

Answer

User-defined triggers

Question 15

You need to verify the integrity of a large file uploaded to GCS. What is the standard way to do this?

Accepted Answer

Check MD5 hash. The standard way to verify integrity of a file uploaded to GCS is to compare its MD5 hash (or CRC32C) against the source file's hash. GCS stores the MD5 and CRC32C checksums as object metadata, and you can compare them to confirm the upload was not corrupted. This is the canonical integrity-verification method.

Answer

Use Cloud Logging

Answer

Check file size

Answer

Use Dataflow