Configuring Fusion Data Foundation

Configure Fusion Data Foundation to set up storage services that support scalable, resilient, and flexible data management for OpenShift® applications.

After the Fusion Data Foundation installation is complete, configure the storage system based on the device type selected during Fusion Data Foundation installation.

Fusion Data Foundation supports the following configuration modes based on device type:
Local
The local mode allows you to deploy Fusion Data Foundation on a smaller scale by using the local storage devices instead of a dedicated storage cluster. It is suitable for edge deployments, testing environments, or smaller setups where a full-scale storage cluster is not required. For instructions, see Configuring Fusion Data Foundation in local mode.
Dynamic
The dynamic mode is designed for environments where existing storage is available. It allows you to leverage an existing storage provider on a OpenShift cluster and build a Fusion Data Foundation cluster or a fault tolerant storage cluster on top of it. This mode integrates with external storage solutions to provide scalable, resilient, and dynamically managed storage resources for your OpenShift cluster. For instructions, see Configuring Fusion Data Foundation in dynamic mode.
External
The external mode is designed for scenarios where the storage infrastructure exists outside the OpenShift cluster. It enables integration with enterprise-grade storage solutions, such as IBM Storage Ceph, IBM FlashSystems, IBM Storage Scale, or IBM Fusion Access for Storage Area Network (SAN), that reside outside the OpenShift environment. For instructions, see Configuring Fusion Data Foundation in external mode.
Provider

The provider mode allows you to run a local Fusion Data Foundation storage cluster that serves as the central storage provider for multiple consumer clusters. It functions similarly to an external Fusion Data Foundation configuration, where the provider cluster acts as the provider and base storage on the host cluster. You can provision new clusters within a managed rack and have them consume storage from the central provider cluster. This mode is available from IBM Fusion 2.8.1 onward with Fusion Data Foundation 4.16. For instructions, see Configuring Fusion Data Foundation in provider mode.

In this mode, Fusion Data Foundation is layered on Red Hat OpenShift Service on bare metal and includes the Fusion Data Foundation provider. IBM Fusion with Fusion Data Foundation supplies the base storage for all Hosted Control Plane-based clusters, as shown in the following figure:

Figure 1. Fusion Data Foundation provider mode
Schematic diagram of Fusion Data Foundation provider mode.

Fusion Data Foundation provider mode is a deployment configuration in which the OpenShift cluster running on bare metal serves local storage for OpenShift virtualization, standard applications, and hosted clusters that are running on top of OpenShift virtualization. Provider mode consists of two key components:

  • Provider - The provider is a deployment mode for Fusion Data Foundation storage services, built on a minimum of three worker nodes using bare metal volumes. This mode allows external clusters to connect through the client operator. Fusion Data Foundation replicates data across these nodes, ensuring resilience and tolerance to node failures.
  • Client - The client is a managed Container Storage Interface (CSI) deployment that connects to the provider to deliver a local storage class for hosted clusters.

From FDF 4.19 onward, provider and internal deployments use the same unified configuration. This allows you to deploy a central OpenShift cluster that provides storage for local workloads, OpenShift virtualization clusters, hosted clusters, and standard clusters. If you plan to use provider mode in the future, ensure that Multus is not enabled. After deployment, you can run the provider cluster locally and add remote clients at a later stage, regardless of where they are hosted.

For instructions on Fusion Data Foundation installation, see Installing Fusion Data Foundation.

After completing the configuration, you can monitor capacity utilization and storage nodes, add nodes, and scale capacity as needed.