AZ-900 Describe Azure architecture and services Practice Question

Answer analysis

Option-by-option breakdown

For each option: why learners choose it and why it is or isn't the right answer here.

• ✗

  Azure Files

  Why it's wrong here

  Azure Files provides fully managed SMB and NFS file shares primarily for lift-and-shift migration and shared access to legacy applications, using standard file protocols rather than an object storage API. Although it supports directories and files, it is not a Hadoop-compatible object store and lacks the massive scale and parallel processing optimizations of a data lake. Its protocol overhead and workload profile are mismatched for big data analytics.

• ✓

  Azure Data Lake Storage Gen2

  Why this is correct

  Azure Data Lake Storage Gen2 (ADLS Gen2) is the correct choice because it combines blob storage's durability and scalability with a hierarchical namespace, enabling true directories and file-level access controls. It exposes a Hadoop-compatible file system through the ABFS driver, allowing Apache Spark, Hive, and other analytics frameworks to run directly on it with high throughput. Fine-grained, POSIX-like ACLs and atomic directory renames make it purpose-built for big data analytics workloads.

• ✗

  Azure Blob Storage (flat namespace)

  Why it's wrong here

  Azure Blob Storage (standard, flat namespace) stores massive amounts of unstructured object data with a flat two-level container/blob naming scheme, lacking a true directory hierarchy. While blobs can back a data lake, tools like Spark and Hive expect HDFS-style directory operations and POSIX permissions; without the hierarchical namespace, operations such as renaming directories become slow and costly. Consequently, standard blob storage alone does not deliver the optimized analytics experience of ADLS Gen2.

• ✗

  Azure Queue Storage

  Why it's wrong here

  Azure Queue Storage is a messaging service for storing and retrieving large numbers of messages to decouple application components; it uses a flat URL-based address space and has no file or directory semantics. It cannot organize data into a hierarchical filesystem, nor does it provide the Hadoop-compatible APIs or ACLs required for big data analytics, making it unsuitable for a data lake.