Object storage concepts
What is object storage?
Object storage stores data as objects in buckets (not as files on a traditional filesystem).
Each object typically has:
- key (name/path-like string)
- data (bytes)
- metadata
Examples: S3-compatible APIs, MinIO, many cloud object stores.
Key differences vs block/file storage
- No POSIX semantics by default (no real directories, no file locks)
- Access via HTTP APIs
- Great for large-scale, durable storage (backups, media, artifacts)
Common concepts
- Buckets: top-level containers
- Prefixes: key naming convention that looks like folders
- Versioning: keep older versions of objects
- Lifecycle policies: tiering, expiration
- Access control: IAM policies, bucket policies
Typical use cases
- Backups and archives
- Static assets and media
- Logs and data lakes
- Build artifacts and container registries