What Is Object Storage
Object storage keeps data as objects inside buckets: each object is the file plus its metadata, addressed by a unique key and reached over an S3-compatible HTTP API instead of a mounted disk. It scales by adding capacity rather than resizing volumes, which makes it the usual home for backups, media libraries and archives.
For site owners and developers choosing where to keep backups and media, and how object storage compares with block and file storage.
knowledge
- Read time
- 3 min
- Sections
- 7
- Updated
- 14 Nov 2026
- Written by
- WebJi® team
Last editorial update: 14 November 2026. Prices follow the dated sources on this page; checkout confirms current charges.
What is object storage?
Object storage keeps data as objects inside buckets: each object is the file plus its metadata, addressed by a unique key in a flat namespace and read or written over an HTTP API rather than a mounted disk. There are no real folders to climb; a key such as backups/2026-11-15/site.tar.gz retrieves the object directly, and the slashes are only part of its name.
Amazon S3, launched in 2006, defined the API most services now follow, so "S3-compatible" has become the common language of object storage, including open-source systems such as Ceph and MinIO. Capacity grows by adding storage nodes to the pool, never by resizing a volume, which is why the same model holds a single WordPress export or a petabyte media archive.
How is object storage different from block and file storage?
Block storage is a raw disk attached to one server, file storage is a shared folder tree mounted over the network, and object storage is a flat pool of self-contained objects reached over HTTP. Choose by workload, not by price per gigabyte.
| Factor | Object | Block | File |
|---|---|---|---|
| Unit | Object plus metadata | Raw volume of blocks | Files in a directory tree |
| Addressing | Key per object | Block offsets on a disk | Paths and folders |
| Access | HTTP API (S3-compatible) | Attached to one server | NFS or SMB share |
| Edits | Replace the whole object | In place, byte by byte | In place |
| Scaling | Very large, add nodes | Fixed volume size | Bound to the file server |
| Bill driven by | GB stored, egress, requests | Provisioned size | Capacity |
| Best for | Backups, media, exports, logs | Databases, VM boot disks | Shared team folders |
For the wider picture of sync drives, backup services and storage types, see what is cloud storage.
Why is object storage used for backups and media?
Object storage suits data written once and read many times, because it scales without resizing, replicates objects for durability, and lets metadata rules manage retention automatically.
- No volume to fill: capacity grows with the archive.
- Durability by replication: objects are copied across devices and often across sites, so one failed disk loses nothing.
- Lifecycle rules: move old snapshots to cheaper tiers or delete them after a set number of days.
- Versioning and object lock: keep earlier versions or make objects undeletable for a period, a defence against ransomware wiping backups.
- HTTP-native: every object can be served by URL, so media can sit behind a CDN.
The trade-off is that editing an object means rewriting it whole, so live databases and application disks stay on block storage.
Which tools work with S3-compatible object storage?
Any tool that speaks the S3 API works with any S3-compatible provider — restic, rclone, the AWS CLI and s3cmd from the command line, and backup tools such as JetBackup and UpdraftPlus from a panel or WordPress. That keeps your backups portable between vendors.
# one-off upload with the AWS CLI to a non-AWS endpoint
aws s3 cp site-backup.tar.gz s3://my-backups/ --endpoint-url https://s3.example-provider.com
# encrypted, deduplicated backups with restic
export AWS_ACCESS_KEY_ID=YOUR_KEY
export AWS_SECRET_ACCESS_KEY=YOUR_SECRET
restic -r s3:https://s3.example-provider.com/my-backups init
restic -r s3:https://s3.example-provider.com/my-backups backup /home/user/public_html
restic -r s3:https://s3.example-provider.com/my-backups snapshotsReplace the endpoint and bucket with your provider's values. Keep the restic repository password somewhere safe; without it the backups cannot be decrypted.
Should object storage be paired with a CDN?
For public media, yes: the bucket acts as the origin and a CDN caches objects at edges near visitors, which speeds delivery and reduces the egress you pay the storage provider. Private backups should never be public, with or without a CDN. How edge caching works is in what is a CDN, and moving WordPress media out to a bucket in offload the WordPress media library.
How is object storage billed?
An object storage bill usually has three parts — gigabytes stored per month, gigabytes delivered out (egress), and small per-request fees — and egress is the line that surprises people. Pulling the same archive out repeatedly can cost more than keeping it.
Compare providers on the storage rate, egress rate, request fees and any minimum storage duration together. WebJi® has no published per-gigabyte object storage rate: bucket work is quoted by email before payment, with 18% GST added. For website backups specifically, WebJi® billing lists the Backup.Help app with 200 GB at ₹1,300 a month before GST, which bundles the software and storage; checkout confirms the current figure. Options are on the object storage page.
How do you back up a website to object storage?
Create a bucket, give a backup tool credentials limited to that bucket, schedule full and incremental backups, test a restore, and set lifecycle rules.
- Create a private bucket in a region you are allowed to store data in.
- Create an access key scoped to that bucket only — never a root or account-wide key.
- Point the backup tool at the S3-compatible endpoint and run the first full backup of files and database.
- Restore to a staging location and confirm the site works; an untested backup is only a hope.
- Enable versioning or object lock, and add lifecycle rules to expire old snapshots.
How often to run each job is covered in how often to back up a website; the managed alternative is cloud server backup.
Frequently asked questions
Can I host a website on object storage?
A static site of HTML, CSS and images can be served from a bucket, ideally behind a CDN. Anything dynamic — PHP, WordPress, a database or a checkout — still needs a hosting server, with the bucket holding media and backups.
Is object storage slower than a disk?
Per request, yes: each read is an HTTP call with more latency than a local disk. Throughput for large files can still be high, and a CDN hides latency for public media. That is why databases stay on block storage.
What is an S3 bucket?
A bucket is a named container for objects, with its own access policy, region and settings such as versioning. "S3 bucket" is used loosely for a bucket on any S3-compatible service, not only Amazon's.
How do I stop a bucket being exposed publicly?
Keep the bucket private by default, block public access at the account or bucket level where the provider supports it, use scoped keys, and share individual files with expiring pre-signed URLs rather than making objects public.
Last editorial update: 15 November 2026. Prices follow the dated sources on this page; checkout confirms current charges.
Ready to get started with WebJi®?
Join 12,000+ businesses across India and overseas. INR pricing, 18% GST shown before payment and 24/7 email support in Hindi and English — average first reply under 60 minutes.