What's new

Each release offers new functions and improvements. IBM is constantly updating the infrastructure, security, and stability of IBM Storage Fusion HCI to improve your experience. Review this information for a high-level summary of the new features and changes in each release.

Note: New IBM Storage Fusion documentation structure:

From the 2.7.x release onwards, the documentation is hosted in separate locations for IBM Storage Fusion and IBM Storage Fusion HCI System variants to cater to audience-specific information.

For more information about what is new in IBM Storage Fusion 2.7.x, see What is new.

To view earlier documentation versions of the product, see https://www.ibm.com/docs/en/storage-fusion.

What is new in 2.7.2

IBM Storage Fusion HCI System 2.7.2 has the following new features:

Next-gen GPU server for watsonx.ai and other AI/ML workloads

A new third generation GPU server is now available for IBM Storage Fusion HCI System. The G03 server supports up to 8 GPUs, enabling large models and more concurrent utilization. You can harness the parallel processing capabilities of G03 server to accelerate AI models training, enhance real-time analytics, and support complex AI-driven tasks with unparalleled speed and efficiency.

You also have the flexibility to add GPU cards from 1 to 8 on G03. Support is available for watsonx.ai 4.8.3 Large Language Models with NVIDIA L40S GPUs. By loading eight L40S GPUs into the G03 server, you can now run all supported watsonx.ai models on IBM Storage Fusion HCI System. With two G03 servers in the rack, a single IBM Storage Fusion HCI System rack can support up to 16 L40S GPUs.

Currency support

What is new in 2.7.1

IBM Storage Fusion HCI System 2.7.1 has the following new features:

Regional DR (Technical preview)

Protect your stateful applications from regional outage impact by asynchronously mirroring application data and etcd metadata between IBM Storage Fusion HCI System clusters that are located in different geographic regions. For more information about Regional DR, see Regional Disaster Recovery - a Technical Preview feature.

Note: Regional DR is released as a technical preview in 2.7.1. Do not use it for the disaster recovery of your mission-critical production applications. Contact IBM support to use any of the IBM Storage Fusion features on a Regional DR setup or to understand what is supported in this technical preview.

Disaster recovery

Starting from 2.7.1, the disaster recovery has a new and improved home in the user interface that provides a rich user experience. The user interface for the disaster recovery has three new subsections:
  • Experience an improved user experience to establish a disaster recovery relationship
  • Manage disaster recovery configuration
  • Monitor disaster recovery health with a new graphical topology to represent disaster recovery health
  • Dedicated applications page to enlist protected applications

For more information about disaster recovery, see Disaster recovery.

Enhanced hardware monitoring

The IBM Storage Fusion HCI System user interface provides an enhanced hardware monitoring experience wherein you can monitor the health of the entire appliance at a glance. You can quickly determine a hardware component in the IBM Storage Fusion HCI System that needs your attention. For more information about hardware monitoring, see Monitoring hardware from IBM Storage Fusion HCI System user interface.

IBM Storage Fusion HCI System hardware

  • New G02 GPU server

    A new server SR650v3 is added to IBM Storage Fusion as a Gen2 GPU server. The G02 provides more powerful GPUs and CPUs for doing inferencing and fine-tuning with larger AI foundation models. For more information about the configuration of the new G02 GPU server, see Hardware configuration SR650 V3 GPU Node (9155-G02).

  • Firmware upgrades

    New firmware versions are available for servers and switches. For the procedure to upgrade, see Upgrading node firmware and Upgrading switch firmware.

Storage for stateful workloads on IBM Cloud Satellite instance

If the IBM Cloud® Satellite instance must provide storage for stateful workloads, then you can use the storage that is provided by IBM Storage Fusion HCI System. For more information about how to implement, see Deploying IBM Cloud Satellite hosts with storage.

New in IBM Storage Fusion Backup & Restore service

  • Keep-alive progress for Backup & Restore jobs

    In prior versions, the Backup & Restore jobs must complete within the configured time out values of the product. Despite the progress, if a job takes longer than the default time out, it fails with timed out error. From this release onwards, you do not have to adjust the time out values for long running jobs. Jobs have a keep-alive timer that keeps the job active as long as it is progressing. As a result, jobs time out only when no progress is made within the time out duration.

  • Support for application VM protection

    Red Hat® OpenShift® VMs that use filesystem-based PVCs can now be backed up and restored just like any other OpenShift workload that runs on a Bare Metal cluster.

  • Monitor and manage IBM Storage Fusion Backup & Restore service capabilities directly from OpenShift Container Platform

    You can monitor and manage Backup & Restore service-related operations from the Red Hat OpenShift Container Platform console. For more information about this feature, see Monitoring and managing Backup & restore service OpenShift Container Platform.

  • IBM Storage Fusion Backup & Restore (Legacy) service support removal from 2.7.1 release

    From this IBM Storage Fusion HCI System release, the Backup & Restore (Legacy) service is no longer supported. For the procedure to migrate to Backup & Restore, see Migrating from Backup & Restore (Legacy). Before you upgrade from earlier versions to this version, go through the Before you begin section of Upgrade prerequisites for Backup & Restore services.

  • Git library of Backup & Restore recipes for assorted workloads
    The IBM Storage Fusion public GitHub repository provides helpful utilities for IBM Storage Fusion. It allows members of the broader IBM Storage Fusion community to share and contribute tools and knowledge.
    • IBM Storage Fusion Backup and Restore recipes
    • Helpful utilities and scripts

      For example, you can run scripts from the command line or command prompt to view recipe workflow logs. You can also view a complete list of backed up or restored resources.

    For the IBM Storage Fusion GitHub repository, see storage-fusion GitHub repository.

    For more information, see IBM Storage Fusion repository for recipes.

  • Resource overwrite flexibility enhancement

    Before 2.7.1 release, by default, resources were not overwritten during a restore operation; however, there was an option to overwrite existing resources. The recipe in 2.7.1 introduces a new field restoreOverwriteResources in the group definition, which provides flexibility to overwrite specific resources during restore.

  • Cancel Jobs

    Support is now available to cancel a job by using a CR. For more information about the cancel job, see Cancel jobs.

New in IBM Storage Fusion Data Cataloging service

The Data Cataloging service includes the following new features in the IBM Storage Fusion 2.7 release:
SD-Monitor

You can now publish more observability metrics, including metrics on Kafka messages. For more information about the SD-Monitor, see SD-Monitor.

FFDC must-gather image

Revamped the log collection for Data Cataloging so that it includes more information that the IBM Support team can use to debug issues. For more information about the FFDC must-gather image, see Collecting logs and metrics.

Python SDK to enable adding non-root applications

The Data Cataloging service provides a new functionality that enables the Python SDK to run non-root applications in cases where non-root access is required. It enables you to run applications without root privileges in OpenShift. For more information about the Python SDK and to enable the addition of non-root applications, see Creating a Data Cataloging application for metadata-based policies.

Bulk adding tags to files through REST API
The Data Cataloging service provides a new feature to add tags to files in a bulk fashion without the need of creating a specific policy for it. With this new capability, you can create a series of tags that are associated to a specific file in a single request with no associated policy interaction. For more information about the bulk adding tags to files through REST API, see /db2whrest/v1/bulk_add_tags/docs: POST.
Improvement of the current data backup and restore strategy

The Data Cataloging service improved the data backup and restore strategy by including a new script. For more information about the improvement of the current data backup and restore strategy, see Backup and restore scripts.

Improved user experience for Single Sign-On (SSO)
Prior to this version, the Single Sign-On (SSO) schema encountered certain limitations in particular deployments that are related to authentication actions. This release includes significant enhancements to address these limitations and improve the user experience, resulting in a more resilient and robust SSO experience.

Serviceability improvements

  • Grafana exporter operator improvements that are related to the Global Data Platform service. For more information about Grafana exporter operator improvements, see Monitoring and logging by using any monitoring tool.
  • Improvements in Log collection:
    • Storage log collection improvement for Global Data Platform and system health package
    • Jobs data is collected as part of all packages for each namespace.

watsonx.ai capabilities

The IBM Storage Fusion HCI System is a hosting platform for watsonx.ai, and you can access watsonx.ai services on the IBM Storage Fusion HCI System. It provides the following capabilities:
  • Leverage the watsonx.ai user interface and REST API interfaces. For more information, see Deploying watsonx.ai capabilities.
  • Schedule regular online backups of a watsonx.ai instance to recover to an earlier point in time in the event of data loss or corruption. For more information, see Deploying watsonx.ai capabilities.
  • Monitor the model serving and tuning pipeline KPIs by using built in OpenShift monitoring capabilities. For more information, see Deploying watsonx.ai capabilities.
  • Leverage the watsonx.ai health and performance monitoring through the IBM Cloud Pak for Data common services. For more information, see Deploying watsonx.ai capabilities.
  • Extend the Backup & Restore service recipes of IBM Storage Fusion HCI System for the watsonx.ai component as well. For more information, see Deploying watsonx.ai capabilities.

Currency support

  • New OpenShift Container Platform version
    IBM Storage Fusion HCI System supports OpenShift Container Platform 4.14.
    Note: It is only supported as an upgrade option and not in fresh installation.
  • New IBM Storage Fusion service versions