Data Engineering Cloud Data Storage Solutions 3 — Questions and Answers
Question 1: Which Azure service provides hierarchical namespace support optimized for big-data analytics?
- Azure Blob Storage (flat)
- Azure Data Lake Storage Gen2 (Correct answer)
- Azure Files
- Azure Queue Storage
Correct answer: Azure Data Lake Storage Gen2
ADLS Gen2 adds a hierarchical namespace on top of Blob Storage for analytics workloads.
Question 2: What is the purpose of partitioning data in cloud object storage for a query engine?
- To encrypt sensitive fields
- To prune irrelevant data and reduce scan volume (Correct answer)
- To enforce schema validation
- To increase replication factor
Correct answer: To prune irrelevant data and reduce scan volume
Partitioning lets query engines skip directories that don't match filter predicates, cutting scan cost.
Question 3: Which consistency model does Amazon S3 now provide for read-after-write on new objects?
- Eventual consistency
- Strong read-after-write consistency (Correct answer)
- No consistency guarantee
- Causal consistency only
Correct answer: Strong read-after-write consistency
S3 provides strong read-after-write consistency for PUTs of new objects.
Question 4: What is a key reason to enable object versioning in a cloud bucket?
- To reduce storage cost
- To protect against accidental overwrites and deletions (Correct answer)
- To speed up queries
- To disable encryption
Correct answer: To protect against accidental overwrites and deletions
Versioning retains previous object copies, allowing recovery from unintended changes.
Question 5: Which open table format adds ACID transactions and time travel on top of a data lake?
- CSV
- Apache Iceberg (Correct answer)
- Plain Parquet
- JSON Lines
Correct answer: Apache Iceberg
Apache Iceberg (like Delta Lake and Hudi) brings ACID transactions and snapshots to lake storage.
Question 6: What does 'data egress cost' refer to in cloud storage pricing?
- Cost to store data at rest
- Cost to transfer data out of the cloud provider's network (Correct answer)
- Cost to encrypt data
- Cost to create buckets
Correct answer: Cost to transfer data out of the cloud provider's network
Egress charges apply when data leaves the provider's network, often to the internet or another region.
Question 7: Which approach minimizes small-file problems in a cloud data lake?
- Writing one file per record
- Compacting many small files into larger files (Correct answer)
- Disabling partitioning entirely
- Storing everything as JSON
Correct answer: Compacting many small files into larger files
Compaction merges many small files into fewer large files, improving read throughput and reducing overhead.
Which Azure service provides hierarchical namespace support optimized for big-data analytics?