About Object Storage for Satellite

IBM Cloud Object Storage for IBM Cloud Satellite offers users the flexibility to run a managed Object Storage service on client-owned on-premises infrastructure, edge locations or third-party public cloud infrastructure.

This introductory offering of Object Storage for Satellite is limited in capabilities and will be expanded on in the future. Keep in mind that not all APIs or connected services may work in the same fashion as Object Storage on IBM Cloud.

Essentially, provisioning an instance of Object Storage for Satellite provides the same familiar interfaces of IBM Cloud Object Storage outside of IBM Cloud.

Object Storage is integrated into Satellite in three different ways: configuration data and backup storage for the Satellite instance itself, as a persistent volume that allows for file-like access, and as a local instance of an IBM Cloud Object Storage service instance. This documentation focuses on the latter - setting up and accessing an instance of Object Storage running on Satellite hardware.

Typical use cases of Object Storage for Satellite

Low latency workloads that need to be run in close proximity to on-premises data and applications including workloads running on factory floors for automated operations in manufacturing, real-time patient diagnosis, and media streaming.

Data residency requirements or those in regulated industries that need to securely store and process customer data that needs to remain on-premises or in locations where there is no public Cloud Object Storage service.

Edge or IOT applications that collect and process data on the edge of network for new workloads from devices and users such as data collection and processing, location-based media, autonomous vehicle data, analytics, and machine data controls for manufacturing.

Hybrid workloads that require management of data between on-premises infrastructure, edge, public cloud or any multi-cloud installation.

How Object Storage for Satellite works

COS on Satellite Architecture — Figure 1. Object Storage for Satellite Architecture

A Satellite administrator needs to configure a new "Location" using the Satellite console and assigns hosts for the Satellite Control Plane.
After the new location is created and accessible, an Object Storage administrator provisions the Object Storage instance in the new location.
The Satellite administrator assigns the appropriate hosts and block storage to the new Object Storage for Satellite cluster.
The new instance is available for both Object Storage bucket configuration and data operations.

Connecting to Object Storage for Satellite

In order to interact with object storage, a client makes API calls to a service endpoint. In a Satellite configuration, these should not be confused with link endpoints which are used for communication between services.

The service endpoint that is used for reading and writing data typically takes the form of https://s3.{cos-instance-uuid}.{location-id}.cloud-object-storage.appdomain.cloud and can be found under the Endpoints section of the object storage console.

COS on Satellite Endpoints — Figure 2. Object Storage for Satellite Endpoints

Keep in mind that requests made to Object Storage for Satellite infrastructure must originate within the satellite location as the service endpoint may not be accessible from the outside of that location.

What features are currently supported?

Any Key Protect instances must be in IBM Cloud and must be located in the same IBM Cloud region from where the Satellite location is managed.

Activity Tracker will log service instance creation and deletion, but not any actions specific to object storage, such as listing buckets or reading/writing data.

Other features that are currently not supported (such as Activity Tracking, Metrics Monitoring, Compliance, Security and Compliance Center) will be added in the future.