What is Object Storage?
Respuesta rápida
Object Storage is S3-compatible bucket storage for files you write once and read many times, backups, static assets, media, archives. You create it as a project in Portal, then connect with any S3-compatible tool pointed at your project's own Worldstream endpoint.
What it's for
Object Storage suits data you write once and read repeatedly rather than data you're constantly editing in place: automated backups, static website assets, media libraries, log archives, and anything you want to serve without running your own file server.
How it works
Data lives in buckets inside an Object Storage project. The project speaks the S3 API, so any S3-compatible tool or SDK, such as the AWS CLI, rclone, or boto3, works against it once it's configured with your project's endpoint and credentials.
To create your first project, open Object Storage in Portal and click Create Project. On the first project you choose a redundancy level and a billing model: Pay per use or Reserved quota. Setup takes roughly 5 to 7 minutes. Once the project's status shows ACTIVE, you manage it through three tabs: Buckets, Usage, and Settings.
Choosing a redundancy level
Two tiers are on offer, both currently in the eu-west-1 region:
| Tier | What it protects against |
|---|---|
| Distributed Storage | Redundant servers, disks and network within one datacenter. Becomes read-only if that datacenter is lost. |
| Geo-Redundant Storage | Multi-site replication. Can survive a full datacenter outage. The premium tier. |
Both tiers are about redundancy, not access speed. There is no cold or archive tier and no Glacier-style storage class, so there is nothing to age data into and nothing to restore before you can read an object. Archiving is a use case you run on these two tiers. Object Storage tiers and lifecycle covers what that means in practice.
Create Project
Choose redundancy level
| Tier | Region |
|---|---|
| Distributed Storage | eu-west-1 |
| Geo-Redundant Storage | eu-west-1 |
Billing model
Access credentials aren't switched on with an "Enable" control. They appear through a reveal control in the portal, or you generate a fresh pair with Rotate credentials.
Worldstream is not Amazon AWS: Object Storage speaks the S3 protocol, but it isn't an AWS account and it isn't reachable at AWS's default endpoints. Tools like the AWS CLI,
rclone, and boto3 must be pointed at your Worldstream S3 endpoint on every single call. Theaws_*parameter names you see in the AWS CLI and boto3, such asaws_access_key_id, are just what those libraries call the fields. They hold your Worldstream keys, not an AWS account.
| Property | Value |
|---|---|
| Protocol | S3-compatible |
| Endpoint pattern | A unique S3 endpoint per project, shown in Portal under the project's Settings tab |
| Credentials | Access Key and Secret Key, shown on the project's Settings tab in Portal (the secret is shown once) |
| Project setup time | Approximately 5 to 7 minutes |
| Storage tiers | Two redundancy classes, Distributed Storage and Geo-Redundant Storage. No cold or archive tier |
| Starting limits | 3 buckets per tier, 20 access keys per project, reserved quota up to 500 TB. The bucket limit is a default an administrator can change. See Object Storage limits |
| Max object size | Not published. The S3-compatible API returns an error naming the limit if you cross it, or open a support ticket for the figure up front |
| Billing model | Pay per use or Reserved quota, chosen when you create the project |
Where to find your own figures: Starting limits, and the limits that are not published at all, are set out in Object Storage limits and quotas explained. For current rates, see worldstream.com or open a support ticket. Your S3 endpoint doesn't follow a fixed hostname pattern either. It's generated per project and shown on that project's Settings tab once the project is active.
Getting started
Create a project from Portal's Object Storage section, choose redundancy and billing, then reveal or rotate your Access Key and Secret Key once the project is ACTIVE. From there, connect with the AWS CLI, rclone, or boto3. See the full walkthrough with configuration examples in the CLI and SDK reference below.