GCP-ADP Data Preparation And Ingestion Practice Test 1 — 15 Questions

Question 1

You need to load CSV files from GCS to BigQuery. The schema changes frequently. Which approach handles schema evolution automatically?

Accepted Answer

BigQuery Load job with auto-detect. BigQuery load jobs with schema auto-detect inspect the CSV header and sample rows to infer column names and types, and when combined with the load job's schema update options they can add new columns automatically as the source schema evolves. This is the only listed approach that adapts to frequent schema changes without code changes. Dataflow, Dataprep, and Cloud Functions all require the schema to be defined or mapped explicitly, so they break when columns are added or renamed.

Answer

Dataprep with rigid recipe

Answer

Cloud Functions with manual schema mapping

Answer

Dataflow with fixed schema

Question 2

You need to ingest small files from an on-premises server to Cloud Storage once daily. Which tool is most cost-effective and simple for this?

Accepted Answer

gsutil rsync. gsutil rsync is the standard tool for simple, scheduled file synchronization between local systems and GCS.

Answer

Cloud Pub/Sub

Answer

Dataflow

Answer

Storage Transfer Service

Question 3

You want to run a Dataflow pipeline on a schedule. What should you use?

Accepted Answer

Cloud Scheduler. Cloud Scheduler is a fully managed cron service that can trigger HTTP endpoints, Pub/Sub topics, or App Engine handlers on a defined schedule. To run a Dataflow pipeline on a schedule, you configure Cloud Scheduler to invoke the Dataflow template launch API (or a Cloud Function that launches it) at the desired intervals. This is the canonical GCP pattern for scheduled batch Dataflow jobs.

Answer

Dataflow Flex Templates

Answer

Cloud Composer

Answer

Cloud Functions

Answer

Pub/Sub

Question 4

You are cleaning data in Dataprep and realize you need to change the format of a date column. Which feature should you use?

Accepted Answer

Recipe Builder. Dataprep's 'Transform' interface allows you to apply functions to columns to change their format.

Answer

Dataflow Job Inspector

Answer

IAM policies

Answer

Cloud Scheduler

Question 5

What is the recommended format for data storage in BigQuery to maximize performance?

Accepted Answer

Avro. Columnar formats like Avro or Parquet are optimized for BigQuery's columnar storage architecture.

Answer

JSON

Answer

XML

Answer

CSV

Question 6

You need to ingest data from a legacy database into Google Cloud. The database has strict network egress controls. How should you connect?

Accepted Answer

Cloud VPN. Cloud VPN is the correct choice because it establishes an encrypted tunnel over the public internet between the on-premises network and Google Cloud, which satisfies the legacy database's strict egress controls without requiring the database to expose itself publicly. It is a standard, supported connectivity option for hybrid ingestion scenarios. The VPN tunnel allows the database to push data to a private GCP endpoint while keeping all traffic encrypted and controlled.

Answer

Identity Aware Proxy

Answer

Public Internet

Answer

Service Account

Question 7

You need to ingest small files into Cloud Storage using a command-line tool. Which command is most appropriate?

Accepted Answer

gcloud storage cp. Both `gcloud storage cp` and `gsutil cp` are correct because they are command-line tools designed for copying files to and from Cloud Storage. `gcloud storage cp` is the modern, recommended tool that offers faster performance and is part of the gcloud CLI, while `gsutil cp` is the legacy tool that still works but is deprecated. For ingesting small files, either command is appropriate, but `gcloud storage cp` is preferred for new projects.

Answer

gsutil mv

Answer

gsutil rsync

Question 8

Which TWO of the following are valid ways to ingest data into BigQuery?

Accepted Answer

Streaming API. Option A, the Streaming API, is correct because BigQuery provides the legacy tabledata.insertAll streaming API and the newer Storage Write API for ingesting individual records in real time with low latency. Option E, Batch Load jobs, is correct because BigQuery supports load jobs that ingest data in bulk from sources such as Google Cloud Storage, local files, or other Google services (e.g., via bq load or the jobs.insert API). Option B (direct copy-paste in UI) is not a supported ingestion mechanism for loading data into BigQuery tables. Option C (FTP upload) is not a BigQuery ingestion method, as BigQuery does not accept FTP as a data source. Option D (SQL INSERT statements) is not a valid bulk ingestion path for BigQuery, since DML INSERT is limited and not intended as a primary data-loading method.

Answer

Direct copy-paste in UI

Answer

FTP upload

Answer

SQL INSERT statements

Question 9

Which TWO of the following are benefits of using Cloud Storage?

Accepted Answer

Global availability. Option B (Global availability) is correct because cloud storage services such as Amazon S3, Azure Blob Storage, and Google Cloud Storage replicate data across multiple regions and availability zones, allowing objects to be accessed from anywhere in the world via HTTP/HTTPS endpoints with low latency. Option D (High durability) is correct because these services are designed for extreme durability—for example, Amazon S3 Standard offers 99.999999999% (11 nines) durability by redundantly storing data across multiple facilities, protecting against hardware failures and data loss. The unmarked options do not belong: automatic virus scanning (A) is not an inherent feature of cloud storage and typically requires separate services like antivirus integrations or malware scanning tools; automatic data compression (C) is not a default benefit, as compression must be applied by the client or via specific configurations; and auto-scaling SQL queries (E) is a database capability (e.g., Amazon Aurora or Azure SQL) rather than a benefit of object/blob storage.

Answer

Automatic virus scanning

Answer

Automatic data compression

Answer

Auto-scaling SQL queries

Question 10

Which THREE of the following are valid BigQuery table types?

Accepted Answer

Native. Option A (Native) is correct because BigQuery's standard, internally managed tables are referred to as native tables, where data is stored in BigQuery's columnar storage. Option B (Partitioned) is correct because partitioning is a valid table type/design in BigQuery, where a table is divided into segments (e.g., by ingestion time, DATE/TIMESTAMP column, or integer range) to improve query performance and cost control. Option E (External) is correct because BigQuery supports external tables that query data stored outside BigQuery, such as in Cloud Storage, BigLake, or other federated sources, without loading it into native storage. Option C (Local) is not a BigQuery table type, as BigQuery is a fully managed, serverless cloud warehouse with no 'local' table concept. Option D (Virtual) is not a BigQuery table type; while BigQuery has logical views and authorized views, these are not categorized as 'virtual' tables in the product's table-type taxonomy.

Answer

Local

Answer

Virtual

Question 11

You need to ingest large amounts of unstructured data into Cloud Storage from an on-premises data center with limited bandwidth. Which service should you choose to ensure the most cost-effective and secure transfer?

Accepted Answer

Storage Transfer Service. Storage Transfer Service is designed for large-scale, online data movement into Cloud Storage from on-premises or other cloud sources. It supports scheduled transfers, bandwidth throttling, and encryption in transit, making it the most cost-effective and secure choice for limited-bandwidth on-premises ingestion. It avoids the cost and complexity of running a custom pipeline.

Answer

Cloud Pub/Sub

Answer

BigQuery Data Transfer Service

Answer

Cloud Dataflow

Question 12

You have a large CSV file in Cloud Storage that needs to be loaded into BigQuery. The file contains a nested JSON structure in one column. How should you best prepare this data?

Accepted Answer

Use a BigQuery load job with autodetect. Using BigQuery's native JSON support or Dataflow allows parsing nested structures during the load process.

Answer

Convert to Parquet before loading

Answer

Upload as a text file and use BigQuery federated queries

Answer

Use Cloud Storage transfer service

Question 13

You are configuring a Cloud Data Fusion pipeline to ingest data from an external SQL database. You need to ensure that only rows modified within the last hour are ingested. Which feature should you use?

Accepted Answer

Change Data Capture (CDC). Change Data Capture (CDC) in Cloud Data Fusion is designed to capture only the rows that have changed since the last pipeline run, using database transaction logs or timestamp-based tracking. For a requirement to ingest only rows modified within the last hour, CDC provides the precise incremental mechanism, avoiding the overhead of scanning the entire table. It supports plugins like the 'SQL Server CDC' or 'Oracle CDC' sources that read change tables or redo logs, enabling near-real-time ingestion. This is the correct feature because it directly addresses incremental, time-bound data extraction.

Answer

Full load with a SQL filter

Answer

Dataflow template overrides

Answer

Batch processing at high frequency

Question 14

Your company ingests a continuously growing stream of application logs into Pub/Sub. Analysts need to query the most recent 30 days of logs in BigQuery with standard SQL, but they do not want to manage any pipeline code or infrastructure. You must implement the ingestion with the least operational overhead. What should you do?

Accepted Answer

Create a BigQuery subscription on the Pub/Sub topic with a write metadata option, and set the topic's message retention to 30 days.. A BigQuery subscription is a native Pub/Sub push target that streams messages straight into a BigQuery table with no code to write or infrastructure to run. Analysts can query the table with standard SQL and filter on the publish-time column added by write metadata to cover the last 30 days. It is the only listed approach that meets the streaming, no-code, low-overhead constraints.

Answer

Schedule a Cloud Scheduler job that invokes a Cloud Function to pull messages from the Pub/Sub topic and insert them into BigQuery.

Answer

Run a Dataflow streaming pipeline that reads from the Pub/Sub topic and writes to BigQuery using a windowed write.

Answer

Export the Pub/Sub topic to a Cloud Storage bucket, then load the files into BigQuery with a recurring load job.

Question 15

You are ingesting a 2 TB CSV file from Cloud Storage into BigQuery using the bq load command. The file has a header row and uses comma delimiters, but some fields contain embedded commas within double quotes. You need to ensure the data loads correctly without errors. What should you specify in your load command?

Accepted Answer

Set --skip_leading_rows=1 and --quote='"'. The load must skip the header row and correctly interpret quoted fields containing commas. Explicitly setting the quote character to double quote ensures that commas inside quotes are not treated as delimiters. While BigQuery defaults to double quote, specifying it removes ambiguity and guarantees correct parsing for files with embedded commas.

Answer

Set --autodetect=true and --field_delimiter=','

Answer

Set --skip_leading_rows=1 and --field_delimiter=','

Answer

Set --allow_quoted_newlines=true and --field_delimiter=','