Prepare and test your skills

Prepare and test your skills

Free sample question with answers and detailed explanation below. The Cloud Digital Leader practice bank has an additional 3,070 practice questions.Try 10 free realistic questions: instant score, no credit card, and no account required.
A retail organization is designing a cloud storage architecture to collect massive volumes of raw data from diverse sources. The solution must support the following requirements:
JSON clickstream logs, and raw open-format files (Parquet and Avro) at scaleWhich storage paradigm and Google Cloud service should the organization implement?
A data lake is a centralized repository designed to ingest, store, and protect massive volumes of raw data in its native format at any scale. On Google Cloud, Cloud Storage serves as the foundational object storage layer for building data lakes. Unlike structured repositories that require data to be cleaned and formatted before storage, a data lake accepts structured, semistructured, and unstructured data directly from source systems.

JSON or CSV logs, and open analytical formats like Parquet, ORC, and Avro.Apache Iceberg.Compared to databases and data warehouses, which enforce schema-on-write and are optimized for structured records, a data lake on Cloud Storage is the only paradigm that natively handles raw unstructured media alongside semistructured logs with maximum ingestion flexibility for machine learning workloads.
The Lakehouse runtime catalog (formerly known as BigLake metastore) is a fully managed, serverless metadata service on Google Cloud. It acts as a centralized metastore that tracks namespaces, table schemas, partition specifications, and file locations—implementing open standards such as the Apache Iceberg REST Catalog API—so multiple query engines can discover and share tables.
metadata.json files and table snapshots rather than storing the underlying images, JSON logs, or Parquet files.Apache Iceberg.This option would be the right choice if the organization already had a data lake on Cloud Storage and needed a serverless metastore to centralize Apache Iceberg table metadata and coordinate concurrent reads and writes across BigQuery and Apache Spark.
When transitioning from a basic data lake to a data lakehouse, organizations combine Cloud Storage (the data layer where raw Parquet, ORC, or Avro files reside) with the Lakehouse runtime catalog (the catalog layer that organizes those files into logical namespaces and tables).
A transactional database—often referred to as an Online Transaction Processing (OLTP) system—is designed to manage real-time operational data using strict relational schemas. AlloyDB for PostgreSQL is a fully managed, PostgreSQL-compatible relational database service on Google Cloud engineered for demanding enterprise transactional workloads.
Parquet or Avro files inside a relational database degrades performance and drives up storage costs significantly.This option would be correct if the retail organization needed a highly available, PostgreSQL-compatible relational database to process real-time customer orders, manage live inventory updates, or combine operational transactions with analytical views.
In a modern cloud data ecosystem, operational databases like AlloyDB for PostgreSQL often act as upstream data sources. Change data capture (CDC) tools like Datastream replicate operational records from the database into a Cloud Storage data lake or BigQuery data warehouse for downstream analysis.
A data warehouse is an enterprise analytical repository designed to store, organize, and query large volumes of structured and semistructured data for business intelligence and reporting. BigQuery is Google Cloud's serverless, highly scalable enterprise data warehouse, and BigQuery native tables store cleansed, structured data in Google-managed columnar storage.
SQL analytics rather than the raw, schema-free landing zone.This option would be the right choice if the organization needed to store cleansed, conformed, and structured business data in a gold consumption layer to power high-concurrency SQL dashboards, Looker visualizations, and ad hoc business intelligence queries.
While BigQuery can query unstructured or open-format data sitting in a data lake by using external tables or Apache Iceberg tables, the underlying raw files themselves must still reside in Cloud Storage, making a data lake the required storage foundation.
Answer them yourself and get your score the moment you finish. No account needed.
Full access includes
Keep the momentum going with these hand-picked practice scenarios
Last updated