220-1101 Virtualization and Cloud Computing 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.

• ✗

  Block storage

  Why it's wrong here

  Block storage splits data into fixed-size blocks, each with a unique address, typically attached to a single server via iSCSI or Fibre Channel. It offers very low latency for random read/write operations, making it ideal for databases and OS volumes, but it does not natively expose HTTP access and requires a filesystem to be layered on top. Scaling block storage to a massive photo collection would demand complex volume management or distributed filesystem overlays, making it both cost-inefficient and operationally heavy for large-scale unstructured data served over the web.

• ✓

  Object storage

  Why this is correct

  Object storage is designed for unstructured data like photos, videos, and documents. It stores each item as an independent object with a unique identifier and rich metadata, and it is accessed through RESTful HTTP/HTTPS APIs (commonly S3-compatible). This architecture allows for automatic horizontal scaling across many nodes, providing essentially unlimited capacity and high durability at low cost. It also offers features like lifecycle policies to transition data to cheaper tiers, making it the optimal choice for migrating a large collection of user-uploaded photos.

• ✗

  File storage

  Why it's wrong here

  File storage organizes data into a hierarchical directory tree and uses protocols like NFS or SMB, which are optimized for local area network access rather than public internet delivery. It relies on a centralized namespace that becomes a scaling bottleneck as the number of files grows, and it typically requires fiddling with file locks and permissions. For millions of user-uploaded photos, file storage does not provide HTTP-native access or automatic scaling, and its per-terabyte cost for unstructured media is generally higher than object storage, making it a poor fit for this workload.

• ✗

  Hot storage tier

  Why it's wrong here

  A hot storage tier is not a distinct storage class; rather, it is a pricing and performance setting applied to an existing storage type, such as object or blob storage. It indicates that data is accessed frequently and thus incurs a higher per-gigabyte fee with lower retrieval costs. Choosing 'hot' does not provide HTTP access, global namespace, or automatic scaling—those characteristics come from the underlying storage technology. Therefore, selecting a hot tier is a cost and latency decision, not a storage architecture choice, and it cannot replace the need for object storage in this migration.

Go deeper

Related to this question