Cloud storage types: object, block and file storage
In short: Cloud providers offer three main kinds of storage. Object storage holds files behind an API, block storage acts as a disk for one machine, and file storage gives shared folders. Each has tiers that trade price for access speed, and each needs encryption, access control and backups.
Three ways to store data
Choosing storage is one of the first design decisions in any cloud project. The choice affects cost, performance and security, so it helps to know how each type behaves.
| Type | How you use it | Examples | Good for |
|---|---|---|---|
| Object | HTTP API, each object has a key and metadata | Amazon S3, Azure Blob Storage, Google Cloud Storage | Backups, logs, images, static websites, data lakes |
| Block | Attached to a VM as a disk, formatted with a file system | Amazon EBS, Azure Managed Disks, Google Persistent Disk | Operating systems, databases needing low latency |
| File | Mounted over NFS or SMB by many machines | Amazon EFS and FSx, Azure Files, Google Filestore | Shared home folders, lift-and-shift apps |
Object storage is not a disk. You cannot edit part of an object; you replace the whole thing. It scales almost without limit and is cheap, which is why it holds most cloud data.
Tiers and lifecycle rules
Object storage comes in tiers. Hot tiers cost more to store but nothing or little to read. Cool and archive tiers cost less to store but charge to retrieve, have minimum storage periods, and archive data can take hours to restore.
Lifecycle rules automate this. A log bucket might keep 30 days hot, move to cool, then archive, and delete after a year. Make sure the deletion date matches your legal retention policy.
Durability versus availability
- Durability is the chance your data is not lost. Providers copy objects across several devices and often several zones, so durability figures are extremely high.
- Availability is the chance you can reach the data right now. It is lower than durability and depends on the tier and redundancy option.
- Block volumes usually live in a single zone. Take snapshots so a zone failure or deleted disk is not the end.
High durability does not protect you from yourself. If you delete or overwrite a file, or ransomware encrypts it, the provider faithfully keeps the damaged copy. Versioning, snapshots and immutable backups protect against that.
Securing storage
- Block public access by default at the account level.
- Grant access with identity policies and short-lived roles, not shared keys.
- Encrypt at rest (on by default with most providers) and consider customer-managed keys for sensitive data.
- Require TLS for all access.
- Turn on versioning and, for backups, object lock or immutability.
- Log data access and review it.
Tip: publicly readable buckets remain a leading cause of cloud data leaks. See securing Amazon S3 for a worked checklist.
Practice
Which storage type would you choose for a VM running a busy SQL database?
Ten web servers need to read and write the same shared folder. Which type fits?
Your data has very high durability. Why do you still need backups?
Key takeaways
- Object storage is addressed by key over an API; block acts as a disk; file gives shared folders.
- Tiers trade storage price against retrieval cost and speed.
- Lifecycle rules move and delete data automatically.
- Durability is not a backup; use versioning, snapshots and immutability.
- Block public access, encrypt, require TLS and log access.
Original content written by Mohamed Fayas for FM TecHub. © 2026 FM TecHub. You're welcome to share a link to this page; please don't copy the text elsewhere without permission.