IBM Fusion services

In the service-based IBM Fusion, choose the features that you want to deploy. You do not have to install all the services at installation time instead enable a feature that is available as a service on-demand basis.

As an administrator, you can install or enable the required services from the user interface, but make sure you meet the resource requirements of individual services.

You can also manage the upgrade of these IBM Fusion services from the user interface. For steps to upgrade in IBM Fusion, see Upgrading IBM Fusion services.
Note: All IBM Fusion services are supported on the Bare Metal platform of IBM Fusion HCI System.

Global Data Platform service

The Global Data Platform (GDP) in IBM Fusion HCI is a unified data management solution that simplifies access, sharing, and management of data across diverse environments. It offers scalable storage optimized for hybrid cloud and AI workloads, ensuring seamless data availability and mobility across multiple locations. It is a software-defined file and object storage that enables organizations to build a global data platform for artificial intelligence (AI), high-performance computing (HPC), advanced analytics, and other demanding workloads.

The Global Data Platform storage type provides the following features:
  • File storage
  • High availability via capacity-efficient erasure coding
  • Metro and regional disaster recovery
  • CSI snapshot support with built-in application consistency
  • Encryption at rest
  • Ability to mount file systems hosted by remote IBM Storage Scale clusters.

Fusion Data Foundation

Fusion Data Foundation as storage in IBM Fusion. You can enable and manage the Fusion Data Foundation services, deploy, and scale up or out the storage cluster. The Fusion Data Foundation storage type provides the following features:
  • Block, file, and object storage
  • High availability through automatic data replication
  • Metro and regional disaster recovery
  • CSI snapshot support
  • Encryption at rest
There are five deployment modes:
Local
The Fusion Data Foundation Local Mode allows you to deploy Data Foundation on a smaller scale by using the local storage devices instead of a dedicated storage cluster. You can use it for edge deployments, testing environments, or smaller setups where you do not need a full-scale storage cluster.
Dynamic
The Fusion Data Foundation dynamic mode is designed for environments where existing storage is available. You can leverage existing storage provider on a OpenShift® cluster and build a Fusion Data Foundation cluster or a fault tolerant storage cluster on top of it. The dynamic mode integrates with external storage solutions to provide scalable, resilient, and dynamically managed storage resources for your OpenShift cluster.
External
The external mode is designed for scenarios where the storage infrastructure is external to the OpenShift cluster itself. You can leverage existing storage systems or storage that is integrated with enterprise-grade storage solutions like IBM Storage Ceph or IBM FlashSystems that reside outside of the OpenShift environment.
Provider
You run a storage cluster locally that can go and serve multiple mode clusters. The provider mode is similar to an external Fusion Data Foundation configuration. It acts as the provider and base storage on the host cluster. You can provision new clusters within a managed rack and consume storage from the central provider cluster. Here, the provider and consumer are in a hosted control plane relationship. This is available from IBM Fusion 2.8.1 onwards with Fusion Data Foundation 4.16.
Consumer
Your hosted OpenShift cluster can consume storage resources of an external provider Fusion Data Foundation cluster.
For more information about Fusion Data Foundation service, see Introduction to Fusion Data Foundation.

Backup & Restore

The Backup & Restore service provides the following features:
  • Policy-driven backup of applications that run on Red Hat® OpenShift
  • Change block detection for Ceph RBD block volumes. It greatly reduces backup times for applications using block mode volumes, including OpenShift Virtualization VMs based on RBD block volumes.
  • Orchestration of application consistent online backups through recipes
  • Multi-cluster Backup & Restore by using a hub and spoke topology
The Backup & Restore provides application-centric backup and data recovery. To know more about the service and its architecture, see Data protection.

Data Cataloging

The Data Cataloging provides data insight for exabyte-scale heterogeneous file, object, backup, and archive storage on premises and in the cloud. The software easily connects to these data sources to rapidly ingest, consolidate, and index metadata for billions of files and objects. For more information about the Data Cataloging service and its architecture, see Data Cataloging.

Content-Aware Storage (CAS)

It leverages natural language processing (NLP) and AI technologies to extract semantic meaning from unstructured data, such as text, charts, graphs, and images. It allows for more efficient data retrieval and integration with AI workflows, particularly in retrieval-augmented generation (RAG) processes. For more information about the CAS service, see Content-Aware Storage (CAS).