Blob Storage

Blob storage services store data as objects (blobs) in flat namespaces organized by containers or buckets. Unlike file systems with directory hierarchies, blob storage treats each object as an independent unit with a unique key, metadata, and the data itself.

Blob storage is designed for data that does not fit well in traditional databases: media files, user uploads, static website assets, log archives, machine learning datasets, and database backups. It offers virtually unlimited storage capacity at low cost per gigabyte.

Key features include configurable redundancy (local, zone, geo-redundant replication), access tiers (hot, cool, archive for cost optimization), lifecycle policies (automatically move or delete old data), and CDN integration for fast global delivery.

In system design, blob storage is the standard answer for "where do we store files?" Key considerations include upload/download patterns (direct upload via presigned URLs vs. proxy through application servers), access control, and cost optimization through tiering.