bloblagring

Bloblagring, often translated as blob storage, is a storage model optimized for unstructured data. In practice, data is stored as discrete objects called blobs within containers or buckets. Each blob has a unique key and carries both the binary payload and associated metadata.

The data model uses a flat namespace; there is no inherent hierarchical filesystem. Some providers present a pseudo-folder view for convenience, but the underlying organization is key-based. Objects are accessed via HTTP-based APIs (primarily REST) using operations such as PUT, GET, DELETE, and LIST. Large blobs can be uploaded in parts, and many services support metadata tagging and content-type information.

Blob storage is characterized by high scalability and durability. Data is typically replicated across multiple facilities or regions to mitigate failures. Many services offer storage classes or tiers that balance cost and access needs, such as hot, cool, and archival options. Features often include versioning to preserve earlier object states, lifecycle policies to move data between tiers or delete it, and immutability or object locking for compliance.

Common use cases include backups and archives, media libraries and asset hosting, content delivery for websites and apps, and as a data lake or storage layer for analytics. While bloblagring provides easy horizontal scaling and cost efficiency for large unstructured datasets, it does not natively expose POSIX-like filesystem semantics and may incur different performance characteristics depending on object size, access patterns, and egress costs.

Security and governance features typically cover encryption at rest and in transit, identity and access management (IAM) policies, bucket/object ACLs, auditing, and optional data loss prevention controls.