Storage: Block Volumes, Object Storage, File Storage, and Archive Tiers
What this domain tests
Storage questions test matching a workload to the right service, then getting the details right: AD scoping, performance tiers, lifecycle rules, and retrieval behaviour. Cost-oriented scenarios often ask you to tier data automatically.
Core rules
- Block volumes are AD-scoped: the volume and the instance must be in the same AD unless you use a replica. Performance tiers are Lower Cost, Balanced, Higher Performance, and Ultra High Performance. Volumes resize online, and backups are incremental and can be copied to another region.
- Shareable (multi-attach) volumes let a cluster of instances attach one volume, but the clustered file system must be cluster-aware.
- Object Storage is regional, organized into buckets inside a tenancy namespace. Bucket names must be unique within that namespace, not globally. Tiers are Standard, Infrequent Access, and Archive, with lifecycle policies moving objects between them.
- Pre-authenticated requests grant time-limited access to an object without IAM credentials.
- File Storage is NFSv3 with a mount target in a subnet; snapshots appear under the
.snapshotsdirectory, and clones, replication, and elastic throughput are available.
Worked example
Log archiving: a lifecycle policy moves objects from Standard to Infrequent Access after 30 days and to Archive after 90 days. Because Archive retrieval is not immediate, anything the application reads during an incident stays in Standard for 30 days; only compliance-retained logs go to Archive.
Traps
Creating a block volume in a different AD from the instance; assuming Archive retrieval is instant; expecting a bucket name to be globally unique; forgetting TCP 2049 rules to the mount target for NFS; confusing the Object Storage Archive tier with the separate Archive Storage service.