Storage and Data Management
Storage spans block (cloud disks), object (OSS), and file (NAS). For ECS disks, know the types (ESSD, SSD, ultra, basic), that you can resize a disk but must also extend the filesystem inside the OS, and that snapshots are incremental and stored independently of the disk. Automatic snapshot policies apply to disks and are a common operational control.
OSS concepts: buckets, objects, storage classes (Standard, IA, Archive, Cold Archive), lifecycle rules, versioning, and cross-region replication. A candidate must understand that lifecycle rules transition or expire objects, and that Archive objects need restoration before access. NAS provides shared file storage mountable by multiple ECS instances; know the NFS/SMB protocol choice and permission groups.
Backup and DR: snapshots protect disks; OSS versioning and replication protect objects; a hybrid backup plan covers both. Traps: assuming a snapshot restores only one disk when it may be a multi-disk consistency group; forgetting that resizing a disk requires an OS-level filesystem expansion; assuming deleting an ECS instance deletes its snapshots (they persist); confusing OSS storage-class minimum storage durations (early deletion incurs charges).
Rule of thumb: "I resized the disk but space didn't increase" → the filesystem was not extended; the block device grew but the partition/filesystem did not.
Worked example: An OSS object moved to Archive cannot be read immediately; a restore request is needed, and the object becomes temporarily accessible.