GCP-ADP Practice Exam 3 — 20 Questions

Question 1

You have a Data Catalog taxonomy that you want to share across multiple projects. How do you achieve this?

Accepted Answer

Define it in a central project and share access. Data Catalog taxonomies are project-scoped resources, so to share one across multiple projects you define it once in a central (host) project and grant other projects' users or service accounts the appropriate Data Catalog IAM roles (e.g., Data Catalog Viewer/Editor) on that taxonomy. This avoids duplication and keeps policy definitions consistent. Cross-project sharing is achieved through IAM bindings, not by relocating the resource.

Answer

Move the taxonomy to the Organization node

Answer

Duplicate the taxonomy in each project

Answer

Use Cloud Resource Manager to link them

Question 2

What is the role of the 'Dataplex Data Steward'?

Accepted Answer

To manage metadata, quality rules, and compliance. A Dataplex Data Steward is responsible for managing metadata, defining and enforcing data quality rules, and ensuring compliance with governance policies across data assets. This role uses Dataplex features like Business Glossary, Data Quality tasks, and Data Catalog to curate and govern data. It is a governance-focused role, not an engineering or analytics role.

Answer

To write SQL queries for the business

Answer

To develop ETL pipelines

Answer

To manage the underlying infrastructure

Question 3

You are configuring a Dataplex Data Quality task. You want to validate that a 'user_id' column has no NULL values. Where do you define this rule?

Accepted Answer

In a YAML configuration file. Dataplex Data Quality tasks are defined using a YAML specification that describes the rules, including non-null constraints on columns. The YAML file is uploaded to a Cloud Storage bucket and referenced when creating the Data Quality task. This approach allows for flexible, version-controlled rule definitions that are decoupled from the underlying data schema.

Answer

In the Cloud Storage metadata

Answer

In the BigQuery table schema

Answer

In the Dataplex Asset definition

Question 4

A data analyst is working in a Vertex AI Workbench notebook. They want to connect to a private BigQuery dataset. How should they authenticate?

Accepted Answer

The instance's attached service account will handle authentication automatically.. Vertex AI Workbench instances are associated with a service account that provides credentials to Google Cloud APIs. When the notebook calls BigQuery, the client library uses the instance's attached service account automatically via the metadata server, so no manual credentials are needed. This is the secure and recommended authentication method.

Answer

Enter your username and password in every query.

Answer

Hardcode the service account key in the notebook.

Answer

Use the 'gcloud auth login' command in every cell.

Question 5

Which component of Cloud Composer acts as the primary interface for managing and triggering DAGs manually?

Accepted Answer

Airflow Web Interface. The Airflow Web Interface (the Airflow UI) is the primary interface in Cloud Composer for viewing, triggering, and managing DAGs manually. It provides DAG views, task logs, graph view, and manual trigger buttons.

Answer

Cloud Storage Browser

Answer

Cloud Shell

Answer

BigQuery Console

Question 6

A financial analyst needs a Looker dashboard tile that shows month-over-month revenue growth as a percentage. The underlying BigQuery table stores completed transactions, and the analyst wants the tile to update without manual intervention each month. Which approach should the analyst use in Looker?

Accepted Answer

Build a table calculation using offset() to compare each month's revenue with the prior month and express the difference as a percentage.. Growth between consecutive periods requires comparing a row with the row before it, which is exactly what offset() provides inside a table calculation. Because the calculation is evaluated when the query runs, the tile reflects current data on every dashboard load with no rebuild schedule. Dividing the month-over-month difference by the prior month's revenue expresses the change as a percentage, matching the requested metric.

Answer

Define a custom dimension using a CASE statement that labels each month as higher or lower than the previous month.

Answer

Create a derived table in Looker's SQL editor that precomputes monthly growth for every historical month and schedule it to rebuild quarterly.

Answer

Add a measure using the percent_of_total type so Looker automatically expresses each month's revenue as a share of the yearly total.

Question 7

You need to run a Workflows execution based on a schedule. Which Google Cloud service is designed to manage this triggering?

Accepted Answer

Cloud Scheduler. Cloud Scheduler is Google Cloud's fully managed cron service, purpose-built to invoke HTTP endpoints, Pub/Sub topics, or App Engine routes on a defined schedule. Workflows exposes an HTTP trigger endpoint, so a Cloud Scheduler job can POST to that endpoint on a cron expression to start executions. This is the canonical, documented pattern for scheduled Workflows runs.

Answer

Eventarc

Answer

Cloud Tasks

Answer

Pub/Sub

Question 8

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 9

Your Cloud Composer DAG includes a task that runs a long-running data export, and you have observed that the task sometimes exceeds the default Airflow worker pod timeout, causing the task to be killed and retried. You want to increase the allowed execution time for this task without affecting other tasks. What should you do?

Accepted Answer

Set the execution_timeout parameter on the specific task in the DAG.. The execution_timeout parameter on an individual task sets the maximum runtime for that task, overriding any default. This is the correct way to extend time for a specific long-running task without altering global settings. Retries, schedule changes, and worker pod timeouts do not provide a targeted per-task timeout adjustment.

Answer

Add a retry_delay and increase retries in the task's default_args.

Answer

Change the DAG's schedule_interval to run less frequently.

Answer

Increase the worker pod timeout in the Cloud Composer environment configuration.

Question 10

A data engineer must ingest change data capture streams from an on-premises PostgreSQL database into BigQuery with minimal latency and without writing custom extraction code. The source table has a primary key that updates frequently. Which Google Cloud service should be used to replicate these changes continuously?

Accepted Answer

Datastream with a PostgreSQL source and a BigQuery destination. Continuous, low-latency change data capture from PostgreSQL into BigQuery without custom code is exactly what Datastream provides, reading the database log and applying changes to BigQuery. Scheduled transfer jobs and timer-driven Dataflow templates operate in batches and miss intermediate changes, while federated queries read at query time rather than replicating, so none deliver the required streaming replication.

Answer

A scheduled BigQuery Data Transfer Service job with a PostgreSQL connector

Answer

A Cloud SQL federated query from BigQuery against the on-premises PostgreSQL instance

Answer

A Dataflow pipeline using the JDBC-to-BigQuery template run on a timer

Question 11

You have a dashboard in Looker Studio and you want to allow users to dynamically change the date range without editing the report. What should you add?

Accepted Answer

Date range control.. A Date range control in Looker Studio is a native chart component that renders an interactive date picker on the report canvas, letting viewers change the time window of all charts bound to that date dimension without editing the report. It is the standard, purpose-built UI element for dynamic date filtering.

Answer

Parameter.

Answer

Calculated field.

Answer

Chart interaction filter.

Question 12

Which THREE of the following are Dataflow windowing types?

Accepted Answer

Fixed. In Google Cloud Dataflow (Apache Beam), the supported windowing types are fixed windows, sliding windows, and session windows, so options A, B, and D are correct. Fixed windows (A) divide a PCollection into non-overlapping, contiguous time intervals of a set duration, such as every 60 seconds. Session windows (B) group elements that arrive close together in time, separated by a gap duration of inactivity, which is useful for user-activity streams. Sliding windows (D) are defined by a window duration and a slide period, allowing windows to overlap, such as a 60-second window that slides every 30 seconds. Binary (C) and Random (E) are not Beam/Dataflow windowing types; no such windowing functions exist in the Dataflow model.

Answer

Binary

Answer

Random

Question 13

You are writing a complex BigQuery query and need to define a temporary table to use later in the same query. What is the most efficient syntax?

Accepted Answer

WITH clause (CTE).. The WITH clause, also known as a Common Table Expression (CTE), defines a temporary named result set that can be referenced later in the same query. It is the most efficient and readable syntax for complex queries because it avoids repeating subqueries and allows modular composition without persisting data.

Answer

Nested subqueries.

Answer

Variable declaration.

Answer

CREATE TABLE AS SELECT.

Question 14

Which THREE steps are involved in an effective Dataflow pipeline development?

Accepted Answer

Deployment. The three correct steps are Deployment (A), Testing (B), and Development (D). Development (D) is where you author the pipeline code using the Apache Beam SDK, defining PTransforms, sources, and sinks before anything can be run. Testing (B) is essential to validate pipeline logic and correctness, typically via unit tests with TestPipeline and DoFnTester, and integration tests against representative data before production. Deployment (A) is the step of running the pipeline on a runner such as Dataflow, whether via the gcloud command, the Dataflow template, or a CI/CD job, to execute it at scale. Automated email alerts (C) are an operational monitoring concern, not a core pipeline development step, and manual data entry (E) is not part of building a Dataflow pipeline at all.

Answer

Automated email alerts

Answer

Manual data entry

Question 15

You are a data analyst at a marketing firm. You need to query a large dataset stored in BigQuery, but you want to minimize query costs. The dataset contains a column named event_time of type TIMESTAMP. Most of your queries filter on event_time for a specific day. What should you do to reduce the amount of data scanned by your queries?

Accepted Answer

Create a partitioned table on the event_time column.. Partitioning the table on event_time enables BigQuery to prune partitions that do not match the filter. When querying for a specific day, only that day's partition is scanned, reducing the amount of data processed and lowering cost. Other methods like LIMIT, wildcard tables, or clustering do not achieve the same cost reduction for time-filtered queries.

Answer

Add a clustered column on event_time to the table.

Answer

Use the LIMIT clause to restrict the number of rows returned.

Answer

Use a wildcard table query to select only the relevant shards.

Question 16

Which TWO of the following strategies help reduce Cloud Composer costs?

Accepted Answer

Delete unused environments. Option C (Delete unused environments) is correct because Cloud Composer charges for the underlying GKE nodes, Cloud SQL database, and web server that run continuously per environment, so removing environments that are no longer needed eliminates those recurring infrastructure costs. Option E (Enable auto-scaling) is correct because Composer's autoscaling dynamically adjusts the number of worker nodes based on the workload's queue demand, allowing the environment to scale down during idle periods and avoid paying for excess capacity. Option A (Keep all historical logs forever) is wrong because retaining logs indefinitely increases Cloud Logging storage costs rather than reducing them. Option B (Use the largest environment size) is wrong because larger environments provision more workers and thus cost more, not less. Option D (Increase scheduler frequency) is wrong because more frequent scheduler runs increase resource consumption and can raise costs instead of lowering them.

Answer

Keep all historical logs forever

Answer

Use the largest environment size

Answer

Increase scheduler frequency

Question 17

You have a Dataplex Lake containing data from two different regions. How does this affect data access?

Accepted Answer

Users must have access to both regional storage resources. A Dataplex Lake can reference assets (Cloud Storage buckets or BigQuery datasets) that reside in different regions, but Dataplex does not move or replicate the data. Access to each asset is governed by the underlying regional storage resource, so users must have permissions and network access to both regions to query data across them. Thus, users must have access to both regional storage resources.

Answer

Users cannot access data across regions

Answer

Data access is automatically merged

Answer

The Lake automatically replicates data

Question 18

You are using Vertex AI Workbench and want to ensure your notebook environment persists after the instance is stopped. Where should you save your work?

Accepted Answer

In a persistent disk attached to the Workbench instance.. Vertex AI Workbench instances use a persistent disk as the boot disk and for user data, so files saved outside the container image persist across stop/start cycles. Saving work in the persistent disk ensures notebooks, datasets, and environment customizations survive instance shutdown. The root directory of the container image and /tmp are ephemeral and are lost when the instance stops.

Answer

In the root directory of the container image.

Answer

In the local /tmp folder.

Answer

In the system memory.

Question 19

You are building a streaming pipeline using Dataflow to process sensor data arriving via Pub/Sub. You need to handle out-of-order data by allowing events to arrive late. Which Dataflow concept must you configure?

Accepted Answer

Watermarks. Watermarks are Dataflow's mechanism for tracking event-time progress in a streaming pipeline. They represent the point in event time up to which the system believes all data has arrived, so configuring an allowed lateness window relative to the watermark lets late-arriving sensor events still be processed correctly. Without watermarks, Dataflow cannot reason about when a window is 'complete' or how long to wait for stragglers.

Answer

Side Inputs

Answer

Partitioning

Answer

GroupByKey

Question 20

You are loading a daily CSV file from Cloud Storage into a BigQuery table. The file contains a header row, uses comma delimiters, and has some rows with extra trailing columns. You want the load to tolerate these malformed rows by skipping them and recording how many were skipped, without failing the whole job. Which load configuration should you use?

Accepted Answer

Set `skipLeadingRows` to 1 and set `maxBadRecords` to a positive value.. `maxBadRecords` is the setting that lets a load job complete while skipping rows that cannot be parsed, and BigQuery reports how many were skipped. Pairing it with `skipLeadingRows` handles the header so the header is not counted as a bad row, giving a load that tolerates the malformed data and still succeeds.

Answer

Set `skipLeadingRows` to 1 and set `allowJaggedRows` to false.

Answer

Set `skipLeadingRows` to 1 and set `ignoreUnknownValues` to true.

Answer

Set `skipLeadingRows` to 1 and enable `allowQuotedNewlines`.