Redp 5688

Front cover
IBM Storage Fusion Product Guide
Jon Herd
Daniel Danner
Axel Koester
Redpaper
IBM Redbooks
IBM Storage Fusion Product Guide
June 2023
REDP-5688-00
Note: Before using this information and the product it supports, read the information in “Notices” on page v.
First Edition (June 2023)
This edition applies to Version 2, Release 5, Modification 2 of IBM Storage Fusion HCI System and IBM
Storage Fusion.
© Copyright International Business Machines Corporation 2023. All rights reserved.

Note to U.S. Government Users Restricted Rights -- Use, duplication or disclosure restricted by GSA ADP Schedule
Contract with IBM Corp.
Contents
Notices . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .v
Trademarks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . vi
Preface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . vii
Authors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . viii
Now you can become a published author, too! . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ix
Comments welcome. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ix
Stay connected to IBM Redbooks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ix
Summary of changes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xi
June 2023, First Edition (minor update) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xi
Chapter 1. Why containers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1

1.1 Why move to containers when VMs seem just fine. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
1.2 Faster time to market in IT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
1.3 Infrastructure as Code and Disaster Resiliency . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
1.4 Cyber Security and Cyber Resiliency . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
1.5 Zero-trust. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
1.6 Workload Scalability . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
1.7 Deconstruction silos with Data Bridging . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
Chapter 2. Why IBM Storage Fusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11

2.1 Why IBM Storage Fusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
2.2 Why Red Hat OpenShift on Bare Metal . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
2.3 Mission Critical Workloads in Containers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
2.4 Getting Workloads into Production faster . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
2.5 Storage for Containers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
2.6 Container Platform Resiliency and High availability . . . . . . . . . . . . . . . . . . . . . . . . . . . 16
2.7 Data Protection for Containers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18
2.8 Red Hat OpenShift made easy: Day2 Operations with IBM Storage Fusion . . . . . . . . 20
Chapter 3. Flavors of IBM Storage Fusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25

3.1 Integration with legacy VM infrastructures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26
3.2 Three Tier Architecture System . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26
3.3 Hyper-Converged Infrastructure (HCI) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27
3.4 IBM Storage Fusion Options . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28
3.4.1 IBM Storage Fusion Software Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28
3.4.2 IBM Storage Fusion HCI System Appliance Overview . . . . . . . . . . . . . . . . . . . . . 29
3.4.3 IBM Storage Fusion HCI System Software Overview . . . . . . . . . . . . . . . . . . . . . . 30
3.4.4 IBM Storage Fusion HCI System Hardware Overview . . . . . . . . . . . . . . . . . . . . . 31
3.4.5 Scalability in HCI . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33
3.4.6 Configuring the IBM Storage Fusion HCI System with IBM Storage Modeller . . . 34
3.4.7 VMware versus bare metal . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35
3.4.8 Use cases for IBM Storage Fusion software. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37
Chapter 4. IBM Storage Fusion Terminology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39

4.1 Operators . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40
4.2 Upgrading . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40
4.3 Management UI. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40
© Copyright IBM Corp. 2023. iii

4.4 IBM Storage Scale Erasure Code Edition (ECE) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40
4.5 Red Hat OpenShift Virtualization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41
4.6 IBM Storage Fusion Data Cataloging service . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41
4.7 IBM Cloud Satellite . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42
4.8 Red Hat Advanced Cluster Management (RHACM) . . . . . . . . . . . . . . . . . . . . . . . . . . . 42
4.9 IBM Storage Protect Plus . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43
4.10 Call Home for all IBM Storage Fusion HCI System. . . . . . . . . . . . . . . . . . . . . . . . . . . 43
4.11 Active File Management (AFM). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43
4.12 AI Support (AI GPU) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44
4.13 Synchronous Storage Replication with Metro-DR . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45
4.14 IBM Storage Fusion Reference Information and Links . . . . . . . . . . . . . . . . . . . . . . . . 46
4.14.1 New features and changes in 2.5.2 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 46
4.14.3 New features and changes in 2.4 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48
4.14.8 IBM Storage Fusion Useful Links . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 51
Related publications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53
IBM Redbooks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53
Online resources . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53
Help from IBM . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 54
iv IBM Storage Fusion Product Guide

Notices
This information was developed for products and services offered in the US. This material might be available
from IBM in other languages. However, you may be required to own a copy of the product or product version in
that language in order to access it.
IBM may not offer the products, services, or features discussed in this document in other countries. Consult
your local IBM representative for information on the products and services currently available in your area. Any
reference to an IBM product, program, or service is not intended to state or imply that only that IBM product,
program, or service may be used. Any functionally equivalent product, program, or service that does not
infringe any IBM intellectual property right may be used instead. However, it is the user’s responsibility to
evaluate and verify the operation of any non-IBM product, program, or service.
IBM may have patents or pending patent applications covering subject matter described in this document. The
furnishing of this document does not grant you any license to these patents. You can send license inquiries, in
writing, to:
IBM Director of Licensing, IBM Corporation, North Castle Drive, MD-NC119, Armonk, NY 10504-1785, US
INTERNATIONAL BUSINESS MACHINES CORPORATION PROVIDES THIS PUBLICATION “AS IS”

WITHOUT WARRANTY OF ANY KIND, EITHER EXPRESS OR IMPLIED, INCLUDING, BUT NOT LIMITED
TO, THE IMPLIED WARRANTIES OF NON-INFRINGEMENT, MERCHANTABILITY OR FITNESS FOR A
PARTICULAR PURPOSE. Some jurisdictions do not allow disclaimer of express or implied warranties in
certain transactions, therefore, this statement may not apply to you.
This information could include technical inaccuracies or typographical errors. Changes are periodically made
to the information herein; these changes will be incorporated in new editions of the publication. IBM may make
improvements and/or changes in the product(s) and/or the program(s) described in this publication at any time
without notice.
Any references in this information to non-IBM websites are provided for convenience only and do not in any
manner serve as an endorsement of those websites. The materials at those websites are not part of the
materials for this IBM product and use of those websites is at your own risk.
IBM may use or distribute any of the information you provide in any way it believes appropriate without
incurring any obligation to you.
The performance data and client examples cited are presented for illustrative purposes only. Actual
performance results may vary depending on specific configurations and operating conditions.
Information concerning non-IBM products was obtained from the suppliers of those products, their published
announcements or other publicly available sources. IBM has not tested those products and cannot confirm the
accuracy of performance, compatibility or any other claims related to non-IBM products. Questions on the
capabilities of non-IBM products should be addressed to the suppliers of those products.
Statements regarding IBM’s future direction or intent are subject to change or withdrawal without notice, and
represent goals and objectives only.
This information contains examples of data and reports used in daily business operations. To illustrate them
as completely as possible, the examples include the names of individuals, companies, brands, and products.
All of these names are fictitious and any similarity to actual people or business enterprises is entirely
coincidental.
COPYRIGHT LICENSE:
This information contains sample application programs in source language, which illustrate programming
techniques on various operating platforms. You may copy, modify, and distribute these sample programs in
any form without payment to IBM, for the purposes of developing, using, marketing or distributing application
programs conforming to the application programming interface for the operating platform for which the sample
programs are written. These examples have not been thoroughly tested under all conditions. IBM, therefore,
cannot guarantee or imply reliability, serviceability, or function of these programs. The sample programs are
provided “AS IS”, without warranty of any kind. IBM shall not be liable for any damages arising out of your use
of the sample programs.
© Copyright IBM Corp. 2023. v

Trademarks
IBM, the IBM logo, and ibm.com are trademarks or registered trademarks of International Business Machines
Corporation, registered in many jurisdictions worldwide. Other product and service names might be
trademarks of IBM or other companies. A current list of IBM trademarks is available on the web at “Copyright
and trademark information” at http://www.ibm.com/legal/copytrade.shtml
The following terms are trademarks or registered trademarks of International Business Machines Corporation,
and might also be trademarks or registered trademarks in other countries.
IBM® IBM Spectrum® Redbooks (logo) ®
IBM Cloud® IBM Spectrum Fusion™ Satellite™
IBM Cloud Pak® Passport Advantage® Spectrum Fusion™
IBM Cloud Satellite® Redbooks®
The following terms are trademarks of other companies:
The registered trademark Linux® is used pursuant to a sublicense from the Linux Foundation, the exclusive
licensee of Linus Torvalds, owner of the mark on a worldwide basis.
Microsoft, and the Windows logo are trademarks of Microsoft Corporation in the United States, other
countries, or both.
Java, and all Java-based trademarks and logos are trademarks or registered trademarks of Oracle and/or its
affiliates.
Red Hat, Ansible, Ceph, OpenShift, are trademarks or registered trademarks of Red Hat, Inc. or its
subsidiaries in the United States and other countries.
VMware, and the VMware logo are registered trademarks or trademarks of VMware, Inc. or its subsidiaries in
the United States and/or other jurisdictions.
Other company, product, or service names may be trademarks or service marks of others.
vi IBM Storage Fusion Product Guide

Preface
This IBM® Redbooks® publication offers a short overview of IBM's integrated environment for
container workloads, IBM Storage Fusion. The product comes in two variants, IBM Storage
Fusion HCI System including all required hardware, and IBM Storage Fusion SDS
(software-only) for deployment in public or private Clouds.
Note: This Redpaper has been updated with the 2.5.1 release only for IBM Storage Fusion
and the 2.5.2 release of IBM Storage Fusion HCI System.
IBM Storage Fusion 2.5.2 includes IBM Storage Fusion Data Foundation 4.12. Fusion Data
Foundation 4.12 and Red Had OpenShift Data Foundation 4.12 both provide the same
data service, and OpenShift Data Foundation is the Red Hat brand of the IBM Storage
Fusion Data Foundation offering. Fusion Data Foundation and OpenShift Data Foundation
functions are the same: the only difference is the digital signature. References and links to
OpenShift Data Foundation remain valid. Fusion Data Foundation 4.12 documentation is
integrated within IBM Storage Fusion 2.5.2 documentation.
Note: IBM Spectrum® Fusion HCI and IBM Spectrum Fusion™ have become IBM Storage
Fusion HCI System and IBM Storage Fusion. This Redpaper edition uses the new IBM
Storage brand names introduced in IBM Storage Fusion 2.5.1.
IBM Spectrum Discover functionality is now available as an integrated service within IBM
Storage Fusion called Data Cataloging.
See Evolving the IBM Storage Portfolio Brand Identity and Strategy to learn more about
how IBM Storage Fusion HCI System and IBM Storage Fusion are key to the IBM Storage
Portfolio.
Chapter 1, “Why containers” on page 1 answers general WHY questions: Why are more
and more workloads deployed as containers? Why are containers beneficial for cyber
security? Why should I care about new development and deployment models involving fewer
staff? Why should I care about containers when VMs seem just fine?
Chapter 2, “Why IBM Storage Fusion” on page 11 focusses on WHY IBM Storage Fusion:
What are the benefits of running containers in a turnkey Red Hat OpenShift cluster within IBM
Storage Fusion? How can the integrated environment help reduce the effort required for
deployment, high availability design, backup implementation and cyber resilience planning?
Chapter 3, “Flavors of IBM Storage Fusion” on page 25 describes flavors of IBM Storage
Fusion implementations, major differences between HCI System (hardware) and SDS
(software) variants, and which choices to make to right-size the product for current needs
while retaining the ability to scale out at a future stage.
Chapter 4, “IBM Storage Fusion Terminology” on page 39 describes the terminology

used for containers environments and more specifically for IBM Storage Fusion. It also covers
some more advanced functions included in the product.
© Copyright IBM Corp. 2023. vii

Authors
This paper was produced by a team of specialists from around the world working with the IBM
Redbooks, Tucson Center.
Jon Herd is a Senior Executive Advocate working for the IBM Technical Lifecycle (TLS)
EMEA Client Care team in Germany. He covers the United Kingdom, Ireland, and Europe,
advising customers on the portfolio of IBM storage products, including the IBM Storage
Fusion range. Jon has been with IBM for more than 45 years, and has held various technical
roles, including Europe, Middle East, and Africa (EMEA) level support on mainframe servers
and technical education development. He has written many IBM Redbooks publications about
IBM Storage and FlashSystem products and is an IBM Redbooks Platinum level author. He
holds IBM certifications in IBM Product Services L3 Thought Leader level, and IBM Technical
Specialist L1 experienced level. He is also a certified Chartered Member of the British
Computer Society (MBCS - CITP), a Certified Member of the Institution of Engineering and
Technology (MIET) and a Certified Technical Specialist of the Open Group.
Daniel Danner is the Lead Architect for IBM Storage Fusion in the D/A/CH region. Before he
joined IBM, he gathered 20 years of experiences working in enterprise IT environments
dealing with mission critical workloads for a large IBM account. Daniel is recognized for his
ability to combine hands on experience, customer view and ability to explain complex
circumstances in a consumable manner. Besides his work on IBM Storage Fusion, he leads
the IBM internal Ansible Automation Community.
Axel Koester Axel Koester, PhD, is Chief Technologist for the EMEA Storage Competence
Center ESCC in Frankfurt, Germany. Besides caring about all most recent storage
technologies and beyond, he frequently speaks at public conferences on technology outlook
topics and their future implications for IT businesses. Currently he also devotes time to
improving Storage performance models for simulation/prediction tools, including models for
IBM Storage Fusion HCI System. Axel is a current member of the board and former chairman
of the IBM TEC Think Tank in D/A/CH. Prior to joining IBM in 2000, Axel helped build the
Large Hadron Collider project at CERN, Geneva, and worked as researcher for the
Fraunhofer Institutes, after having worked as an object-oriented programming pioneer in the
privately held company Gesytec mbH. Axel holds a PhD in electrical engineering and
computer science from the RWTH Aachen University.
Thanks to the following people for their contributions to this project:
Larry Coyne
IBM Redbooks, Tucson Center
Josh Blumert
IBM Global Sales
Chrisilia Bristow, Matt Divito, Lee Helgeson, Khanh Ngo, Tom O’Brien, Fernando de la Tejera
Ruiz, David Wright
IBM Systems
viii IBM Storage Fusion Product Guide

Now you can become a published author, too!
Here’s an opportunity to spotlight your skills, grow your career, and become a published
author—all at the same time! Join an IBM Redbooks residency project and help write a book
in your area of expertise, while honing your experience using leading-edge technologies. Your
efforts will help to increase product acceptance and customer satisfaction, as you expand
your network of technical contacts and relationships. Residencies run from two to six weeks
in length, and you can participate either in person or as a remote resident working from your
home base.
Find out more about the residency program, browse the residency index, and apply online at:
ibm.com/redbooks/residencies.html
Comments welcome
Your comments are important to us!
We want our papers to be as helpful as possible. Send us your comments about this paper or
other IBM Redbooks publications in one of the following ways:
򐂰 Use the online Contact us review Redbooks form found at:
ibm.com/redbooks
򐂰 Send your comments in an email to:
[email protected]
򐂰 Mail your comments to:
IBM Corporation, IBM Redbooks
Dept. HYTD Mail Station P099
2455 South Road
Poughkeepsie, NY 12601-5400
Stay connected to IBM Redbooks

򐂰 Find us on LinkedIn:
http://www.linkedin.com/groups?home=&gid=2130806
򐂰 Explore new Redbooks publications, residencies, and workshops with the IBM Redbooks
weekly newsletter:
https://www.redbooks.ibm.com/Redbooks.nsf/subscribe?OpenForm
򐂰 Stay current on recent Redbooks publications with RSS Feeds:
http://www.redbooks.ibm.com/rss.html
Preface ix
x IBM Storage Fusion Product Guide
Summary of changes
This section describes the technical changes made in this edition of the paper and in previous
editions. This edition might also include minor corrections and editorial changes that are not
identified.
Summary of Changes
for IBM Storage Fusion Product Guide
as created or updated on June 22, 2023.
June 2023, First Edition (minor update)

This revision includes the following changed information.
Changed information
򐂰 Updated branding information for IBM Storage Fusion Data Foundation (FDF) and Red
Hat OpenShift Data Foundation (ODF). See the IBM Storage Fusion Data Foundation
section in “Scaling Storage for Applications” on page 7 for additional information.
򐂰 Updated storage product names in Figure 3-4 on page 30.
© Copyright IBM Corp. 2023. xi

xii IBM Storage Fusion Product Guide
1
Chapter 1. Why containers

This chapter answers the general WHY questions: Why are more and more workloads
deployed as containers? Why are containers beneficial for cyber security? Why should I care
about new development and deployment models involving fewer staff? Why should I care
about containers when VMs seem just fine?
© Copyright IBM Corp. 2023. 1

1.1 Why move to containers when VMs seem just fine
“I'm too busy chopping wood to sharpen the axe.” says the lumberjack when he is
overwhelmed with workload. It's a common situation in IT as well: Security concepts turn out
to be less secure than expected; disaster protection doesn't protect enough from new types of
disaster that were just discovered. Most of this can be traced back to a lack of time spent on
these challenges, while too much time is spent on “mundane” tasks that could be automated
if they weren't so diverse.
The metaphor of “pets vs. cattle” (commonly attributed to Bill Baker, then of Microsoft) is well
known in that context: To make automation easier and automatic procedures less diverse, do
not treat computers as individual devices, do not give them individual names, do not manage
their IP addresses, do not manually assign workloads, and do not make machines “opaque”
just to isolate dependencies.
Do you think that virtualization achieves that? Think again. Virtualization improves the
efficient use of resources, confines dependencies, and makes workloads easily movable. But:
Each VM is effectively another individual machine, and their count keeps growing. Each VM
demands (at least some) individual care and effort. Hordes of pets (Figure 1-1).
Figure 1-1 Pets versus Cattle1
Can we skip the concept of machines in IT altogether? Yes: This is sometimes referred to
as serverless operations, even though there is still real hardware involved. Managing “cattle”
means more than giving these machines numbered names: The idea of containerization is to
get rid of the machine concept as individual entity. Individual worker nodes or containers do
not need personal attention. Their instantiation, their workload assignment, their termination,
their whole lifecycle is fully automated and driven by short-term demand.
Likewise, security in individual pods or containers won't need personal attention. This is
handled at a higher abstraction level. And nobody looks after disaster resilience in individual
containers. This, too, is usually handled at a higher level.
Eliminating the effort spent on monitoring and managing individual entities leaves us more
time to “sharpen the axe”. The goal of Red Hat OpenShift inside IBM Storage Fusion is not
merely to ease the management of individual containers, but to actually make us forget that
they exist. In contrast, hypervisor environments usually provide tools to manage many pets -
but each of them requires individual care. See Figure 1-2 on page 31.
1 images: creative commons CC-0
2 IBM Storage Fusion Product Guide

Figure 1-2 Managing individual entities versus serverless operations
So, are VMs totally eliminated in a container environment? No: VMs can be very useful to set
up prototypes, to run quick deployment tests, or to add flexibility to control plane nodes. But
for the workload-carrying production worker nodes, VM constructs are obsolete. IBM Storage
Fusion is not built on top of hypervisor technology, even though it has the capability to operate
VMs alongside with containers. Refer to 3.4.7, “VMware versus bare metal” on page 35.
1.2 Faster time to market in IT

With modern continuous delivery methodologies, software release cycles have become very
short. To keep up with that pace, an easier transition from dev-to-test and from
test-to-production is compulsory. Software development itself has also transitioned from
formerly mostly “in-house” to nowadays mostly “re-use” of open-source libraries and
frameworks.
A real-world case: Medical support platform in 3 weeks

During the Coronavirus pandemic outbreak, medical support projects had to be implemented
urgently, with no ready-to-deploy commercial offers available: A newly built Coronavirus test
center in Mainz, Germany, asked IBM to design and build a tailored IT solution for secure
appointment logistics and communication among the medical staff. Several thousand patients
were expected within the first days of operation. Leveraging container technology and prebuilt
service templates from the IBM Cloud®, the hired IBM consulting team was able to roll out a
minimum viable product on-site within weeks, including hardware and network installation.
The final solution grew to a containerized services platform with appointment workflow and
time slot management, handling of digital certificates for testing and vaccination, a patient
survey, a patient chat bot, and a secure messaging app for healthcare workers, all of this
realized by porting code developed on (or for) the IBM Cloud.
The container advantage

In the container world, re-use of preconfigured services is the norm. Containers allow fast
time-to-market due to higher levels of standardization, and easier rollbacks in case of glitches.
Separate environments for dev/test, pilot and production are not obligatory, as the isolation
between instances is handled at an abstract level. Once ready, select parts of the staging pilot
can be declared as production instance - without redeployment, and with a quick undo option.
Moreover, partial updates can be tested against a limited set of end users first:
Let's increase user acceptance by testing modifications at scale (not just at home) with an
increasing number of users, by controlling which range of IP addresses sees which
combination of updates.
We are so much used to seamless app updates on our mobile devices that we keep asking
ourselves: What's holding back larger IT solutions to do the same? The answer is often “weak
dependency management” and “monolithic architecture”. Monoliths, usually large
conglomerates of compiled software, can only be rolled out in one piece. All-or-nothing.
Chapter 1. Why containers 3

In contrast, container-based applications are split into many functional parts (sometimes
down to microservices) bound together at runtime via a dependency description, the recipe.
The recipe elements control which version of which component is part of a production
instance, a progressive deployment instance, beta test instances, and developer instances.
All of these can coexist in the same environment, with containers providing the sandbox
technology to keep data secure.
IBM Cloud Paks

IBM Cloud Paks are prime examples of previously monolithic software that was split into
functional containers and then bundled so that each service can serve multiple purposes.
When configuring Cloud Paks, a dependency checker tells us which services are missing and
adds them to the basket. The more software we add to the shopping list, the smaller the
increments become, as most required services are already included. This is no different from
the way Linux and Unix package managers operate, albeit at a larger scale. We're trading
individual installations for a framework that controls all dependencies and thus significantly
speeds up all future deployments.
Swift deployment of IBM Cloud Paks on IBM Storage Fusion

When planning IBM Storage Fusion for IBM Cloud Pak® workloads, we recommend
estimating the container workload in the appropriate tool, the Cloud Pak sales configurator,
and within minutes obtain an approximation of the required hardware. Use this estimate to lay
out a right-sized IBM Storage Fusion configuration using the Storage Modeller (StorM) tool.
See the example hardware estimation in Figure 1-3.
https://app.ibmsalesconfigurator.com (requires an IBM account and a public IBMid)
Figure 1-3 Hardware estimation for a planned Cloud Pak workload (from ibmsalesconfigurator.com)
Details on configuring IBM Storage Fusion in the StorM tool can be found at 3.4.6,
“Configuring the IBM Storage Fusion HCI System with IBM Storage Modeller” on page 34.

1.3 Infrastructure as Code and Disaster Resiliency
Hardware and network requirements for an application suite, including access to local storage
and compute resources, also constitute dependency chains. Therefore, we can add them to
our “recipe”. The recipe becomes the centerpiece of all containerized application suites,
telling the automation how to select hardware candidates, configure the network, apply
mandatory operating system patches, fetch required services templates from repositories,
and start user-interfacing services in the right order. The recipe is documentation and
automation script in one.
The concept of “describing all in one place” is beneficial when it comes to

򐂰 scaling an application (which services do I need to scale in the backend? On which flavor
of hardware?)
򐂰 restoring a whole environment after an attack or a disaster (can I check the integrity of all
components? In which order should I restart everything?)
In a containerized environment, these two subjects are barely distinguishable.
Building Infrastructure with Developer Tools

Hardware selection instructions in our recipe can be managed in a git repository - with
branching, merging, and roll-back options. The ultimate goal here is “infrastructure as code”,
where hardware selection and network configuration are written by DevOps programmers,
with access to efficient developer pipeline tools. Hardware rollbacks in this context are trivial.
The deployment process becomes increasingly hardware agnostic or “Cloud-inspired”,

making it moveable between public and private clouds, managed clouds services, and
home-cloud locations like IBM Storage Fusion.
1.4 Cyber Security and Cyber Resiliency

Cyber security can be an inherent part of a recipe-based deployment: Just like Linux
packages that are thoroughly checked for their integrity before being considered for
deployment, complex application components (and even hardware) can be checked for
integrity before going live. To avoid upstream supply chain attacks, fingerprinting and source
authentication are mandatory from the very source of each piece of code.
Apache Log4j lessons learned

Even authenticated and well-known code can have zero-day vulnerabilities that eventually
lead to exploits. In the case of Log4j (a core component in the logging infrastructure of Java
apps) the vulnerability was tied to remote code execution in an unintended way. Patches
became available shortly after this vulnerability was made public, yet many computers
remained unpatched for weeks or even months.
The problem here was weak dependency management and slow update policies: In order to
know which systems/instances need the patch, their reliance on Log4j must be documented.
While easily discoverable in the Log4j case, dependencies will be harder to detect in others -
especially when documentation quality depends on human meticulousness. In a container
environment, this dependency is coded as part of the construction recipe of our application
suite and therefore always up to date.

How containerization improves Cyber Protection
Cyber security can be vastly improved by eliminating the human factor in dependency
management. In a containerized environment, the build process of any containerized
application suite (using the recipe) yields a complete documentation: What has been built, is
documented - and that documentation can also cover server and network configurations, all in
one place. This makes automatic cyber threat scanning and the discovery of indicators of
compromise much easier and streamlined.
Cyber resiliency on the other hand can be improved by automating a recovery process and
testing it regularly, e.g. an application restart in a failover location that has access to a copy of
the most recent data. Testing it regularly is the limiting element here, comparable to sharping
the axe when you should be chopping wood.
Containerized environments simplify frequent recovery testing by design: There is no

difference between restarting and relocating a container, or scaling up a containerized service
by adding instances, which may happen during automatic workload optimization.
Infrastructure-as-code simplifies the deployment in another location with a compatible control

plane and access to the original data. Tools for mirroring or replicating the user data, plus the
ability to restart an application suite elsewhere, are integral parts of IBM Storage Fusion.
Refer to 4.8, “Red Hat Advanced Cluster Management (RHACM)” on page 42 and 4.13,
“Synchronous Storage Replication with Metro-DR” on page 45.
1.5 Zero-trust
On top of typical security procedures, zero-trust means that no “truth” is taken for granted
without proof, and no proof may exceed a certain age.
Since distributed applications can include many services, it makes sense to design a mutual
authentication scheme, especially so when some of these services are externalized. Between
authenticated services, communication can be encrypted in a way that only the intended
receiver can decode the content. See Figure 1-4.
Figure 1-4 Microservice mesh with direct P2P communication (left) and with “zero trust” mutual
authentication (right)
Zero-trust for dynamic environments

An example for a service that scales automatically by adding instances on demand is the user
management component of a streaming platform. To avoid man-in-the-middle attacks, all
parts in this multi-instance communication cosmos must individually authenticate - none is
accepted as trustworthy per se. Container platforms can enforce TLS-based secure
communication between all components of a containerized application suite, making
eaves-dropping and spoofing much harder.

Beyond reducing the risk of attacks, the flow of data and commands can be monitored and
analyzed in machine learning systems, which can throw alerts upon sensing irregularities.
These could be components that never communicated in the past, suspicious service access
patterns, suspicious denials, any others.
1.6 Workload Scalability

A prerequisite for automatic workload scaling is that services must be multi-instance capable.
In complex applications, only those components that consume more CPU than allowed (or
generate more IOs than allowed) need to be scaled. A typical scaling rule could be:
1. Add instance when active instance has >50% CPU utilization.
2. Destroy instance when it shows <10% average CPU utilization and there is more than one
instance left.
3. Start one instance when there is none.
In fact, this automation is not scripted as a sequence of actions, but as a series of outcomes:
An application architect could define that “no instance should continually consume more than
50% or less than 10% CPU” and the workload scheduler component underneath IBM Storage
Fusion (Red Hat OpenShift / Kubernetes) makes it happen, preferring the least utilized
hardware for spreading the workload. Additional rules for throughput or latency optimization
are possible. See Table 1-1.
Table 1-1 Classic versus containerized performance management approach

Classic performance management Containerized performance management
approach approach
When an application exceeds the resources of When a container exceeds 50% CPU utilization
a VM, manually increase the assigned (configurable), the platform starts an additional
resources if appropriate. instance of it.
When all VMs on a system exceed that When an instance consumes less than 10%
system's capabilities, manually migrate the (configurable), the platform destroys it unless it is
least important VMs onto new hosts. the last instance.
Programmers of multi-instance services do not need to re-invent proprietary service APIs and
protocols susceptible to new security flaws: Open-Source solutions exist for every flavor of
workload distribution. Most popular are publisher/subscriber schemas like those proposed in
Apache Kafka, a core component in many scalable apps.
Scaling Storage for Applications

IBM Storage Fusion is leveraging the Red Hat OpenShift Storage class concept that
differentiates between ephemeral local data and persistent data that survives container
restarts. Therefore, persistent data is always kept on shared storage. Scalability for shared
storage means:
򐂰 enough physical storage to fulfill growing capacity and throughput requirements, plus the
ability to increase in-rack physical capacity on the fly.
򐂰 an automatic increase of capacity assigned to an application (within configurable limits)
and a dynamic sizing of associated services like caching, mirroring, redundancy
monitoring, or access control.

Depending on the flavor of distributed storage implemented, associated procedures will look
slightly different, even though IBM Storage Fusion will hide most of this to the administrator
and make it invisible for a code developer, who will only have to choose between persistent
storage classes offered.
The two supported storage options are, at the time of writing:
IBM Storage Fusion Data Foundation: IBM Storage Fusion 2.5.2 includes IBM Storage
Fusion Data Foundation 4.12. Fusion Data Foundation 4.12 and Red Had OpenShift Data
Foundation 4.12 both provide the same data service, and OpenShift Data Foundation is the
Red Hat brand of the IBM Storage Fusion Data Foundation offering. Fusion Data Foundation
and OpenShift Data Foundation functions are the same: the only difference is the digital
signature. References and links to OpenShift Data Foundation remain valid.
Fusion Data Foundation 4.12 documentation is integrated within IBM Storage Fusion 2.5.2
documentation. IBM Storage Fusion Data Foundation is based on the open-source projects
Ceph, Rook, and Noobaa, and provides a variety of access protocols while being able to use
a wide range of physical storage sources. Redundancy is achieved by storing three copies,
i.e. the erasure coding feature of Ceph is not supported by IBM Storage Fusion Data
Foundation.
ECE: Erasure Coding Edition of IBM Storage Scale, delivered fully integrated with IBM
Storage Fusion HCI System hyperconverged infrastructure. Local node SSDs are combined
into a dispersed, “write-many” persistent storage layer. In addition, IBM Storage Scale ECE
supports multiple access protocols, advanced automated file management (AFM), rule-based
data placement and relocation, geographically dispersed caching, metro replication, and
much more. IBM Storage Scale aims at bottleneck-free parallel scalability by design. Data
redundancy is achieved by erasure coding with high 64% usable-to-raw storage capacity
efficiency when protecting against double failures. This is approximately 2× the capacity
efficiency which can be achieved with three copies of data (33%).
Scaling Backup
IBM Storage Fusion provides a platform-independent backup based on IBM Storage Protect
Plus, where data can be protected on air-gapped storage services and/or devices outside of
the production cluster. This makes sure that data which is being restored after a major
incident is not dependent on the original source of the data, as that may have been
compromised during the incident.
While all backup components inside IBM Storage Fusion scale on demand, externalized
components (like moving copies onto an air-gapped tape library or object storage) must be
managed and scaled individually, since Red Hat OpenShift is not in charge. On the other
hand, previously existing compatible backup environments and procedures can be re-used or
integrated.
1.7 Deconstruction silos with Data Bridging

Data silos limit the efficiency of many implementations, be it for consistent automation,
security, or shared resource usage. Choosing ECE and therefore IBM Storage Scale as a
storage layer in IBM Storage Fusion HCI System helps break up silos and facilitates a single
logical view of data across platforms, rather than having to deal with “data tourism”
(channeling data between various services and apps, usually generating lots of copies).
IBM Storage Scale can:

1. make user data persistent, i.e. avoid deletion when a service instance is terminated.

2. make user data cluster-wide persistent, so that containers can be restarted on any node at
equal cost, with immediate access to their persistent data.
3. harden user data by dispersing it over several storage-rich nodes with space-efficient
erasure coding, as explained previously in “Scaling Storage for Applications” (ECE:
Erasure Coding Edition).
4. make user data persistent across clusters through metro mirroring, as described in 4.8,
“Red Hat Advanced Cluster Management (RHACM)” on page 42 and 4.13, “Synchronous
Storage Replication with Metro-DR” on page 45.
5. allow asynchronous access to data through other protocols like Object S3 / NFS / etc.,
either for data ingest or for use in legacy applications.
6. provide data protection connectors to the backup component IBM Storage Protect Plus.
7. avoid “cabled security” (solely based on hardware/network isolation) while providing
excellent protection, which helps prevent silos.
Companies using IBM Storage Scale to its full extent typically refer to it as an enterprise data
highway, data bus, or hybrid data bridge. It shortens time-to-market by removing the need to
implement complex and custom data sharing mechanisms between new and existing data
sources. Instead, IBM Storage Scale creates a service bridge which easily and transparently
spans the distance using access credentials and existing storage protocols such as NFS and
S3 Object.
IBM Storage Fusion HCI System supports various interaction mechanisms with other
clusters, applications, or other data sources:
򐂰 Cross-cluster mounts to other IBM Storage Fusion or IBM Storage Scale clusters
(roadmap)
򐂰 Data source for distant caches (AFM - Active File Management source)
򐂰 Data cache from a remote AFM source
򐂰 Scheduled asynchronous replication, RSYNC
򐂰 Synchronous replication, Metro-DR, supporting disaster recovery in a different site
򐂰 Import from an IBM Cloud Object Storage system while sharing its name space.

2
Chapter 2. Why IBM Storage Fusion

This chapter focusses on WHY IBM Storage Fusion: What are the benefits of running
containers in a turnkey Red Hat OpenShift cluster within IBM Storage Fusion? How can the
integrated environment help reduce the effort required for deployment, high availability
design, backup implementation and cyber resilience planning?

2.1 Why IBM Storage Fusion
IBM Storage Fusion HCI System is IBM “bare metal” Red Hat OpenShift cluster in-a-box that
removes the infrastructure challenge of bare metal Red Hat OpenShift while enjoying its
benefits. The complete hardware and software stack is supported by IBM and contains data
protection and resilience capabilities as well as an enterprise class high performance storage.
IBM Storage Fusion SDS takes this software stack to custom-built Red Hat OpenShift
clusters in the private and public cloud.
In the following sections we will explore these benefits in more detail.
2.2 Why Red Hat OpenShift on Bare Metal

IBM Storage Fusion HCI System has been designed based on bare metal Red Hat OpenShift
due to its many advantages over hypervisor-based Red Hat OpenShift. Let's have a closer
look at two of them.
Lower Total Cost of Ownership (TCO)

Red Hat OpenShift subscriptions (or licenses) for bare metal are counted by processor
sockets: One subscription covers 2 sockets with a total of up to 64 CPU cores. Red Hat
OpenShift on a hypervisor or any other platform is licensed by CPU cores assigned to the
worker nodes, which usually turns out more expensive.
Taking a brief look at the list prices highlights the benefit:
A 32 core worker node licensed with bare metal subscriptions reduces the Red Hat OpenShift
license cost by 81%. Considering our 64-core node, the savings grow to 91%. See
Figure 2-1.
Figure 2-1 Comparison to show increased savings using 64 core node
This significantly lowers the TCO of the container platform. The street price of the smallest
HCI System configuration with three 32 core worker nodes (including all licenses), is lower
than the price of the software licenses for a hypervisor-based cluster, even without taking the
hardware and hypervisor cost into the equation.
Removing the Hypervisor from the equation

The Hypervisor is a software layer that not only needs to be licensed, it also needs to be
installed, managed and updated, for which large IT shops usually employ a dedicated team of
specialists.
Besides, a Hypervisor requires CPU and memory resources from the host system which
cannot be used by Red Hat OpenShift or the containerized workloads.
The hypervisor will also introduce higher latencies to network and disk IOs that run through its
virtualization layer, which in some cases may impact the application performance. See
comparison in Figure 2-2.

Figure 2-2 Comparison of virtualized infrastructure layers with bare metal on IBM Storage Fusion HCI
System
2.3 Mission Critical Workloads in Containers

Workloads deployed in containers call for a cloud-like operating model:
They should be resilient to restarts and they should be able to scale out by running multiple
instances. Webservers and the famous LAMP stack are considered being low hanging fruit
here: The migration is trivial.
But mission critical monolithic workloads like RDBMS databases can just as well benefit from
the container deployment and operating model. These workloads require the same level of
performance, security, resilience, and data protection as they were previously getting in VMs
or on bare metal servers. These requirements towards the container platform necessitate a
well thought-through architecture and operating model.
IBM Storage Fusion HCI System was built from the ground up for mission critical applications
like for example the IBM Cloud Paks. Its global data platform is based on a high-performance
parallel file system that has a dedicated 200 Gbit network connection per node in order to
avoid bottlenecks.
While the data is encrypted at rest and in flight, it's also resilient to double failures, which are
rectified by the erasure coding algorithm that builds parities spread across multiple nodes.
Chapter 2. Why IBM Storage Fusion 13

Backups are not only taken via snapshots that reside on the file system and can be restored
within minutes, they are as well off-loaded towards S3 storage to create a logical air-gap. The
built-in Metro-DR option creates a high availability concept across two availability zones within
a metropolitan area.
2.4 Getting Workloads into Production faster

Smaller companies are often very agile in building their container platform, as there are fewer
and smaller teams involved. In many cases, only a handful of experts design and operate the
full stack.
Big enterprises are facing different challenges, as multiple teams are involved. Infrastructure,
Storage, Backup, Monitoring, and Operations; each team has its own concept and favorites,
which will bring multiple products and vendors to the table. A heterogeneous complex
software stack is created, with interoperability dependencies and support matrixes from
multiple vendors.
This software stack needs to be monitored and updated, which makes “day2” operations
difficult. Finally, the deployment and configuration of the stack should be automated with
“recipe-style” infrastructure automation in order to avoid mistakes and improve repeatability.
Building a production-ready container platform this way will result in a long-running project
with many resources.
IBM Storage Fusion HCI System is denominated as a “turnkey appliance” because it delivers
a production-ready Red Hat OpenShift cluster in just one day. The appliance is delivered as a
rack with all components mounted, cabled, and tested. Once connected to power and
network, the fully automated installation starts.
Day2 Operations including updates and support activities are made easy as this software
stack is fully supported out of one hand:
IBM provides support for

򐂰 Red Hat OpenShift
򐂰 Global Data Platform (SDS)
򐂰 Data protection
򐂰 Hardware support including Firmware updates (HCI only)
2.5 Storage for Containers

This section describes the advantages of container scalability and resiliency.
Scalability
A container platform is often built and sized for its initial use case. Once the container
adoption in an enterprise gains momentum, new applications with new requirements get
onboarded on the platform and an expansion is needed.
One of the big advantages of a container platform is that compute resources can be easily
scaled out by adding more worker nodes to the cluster.
But what about storage?

Containers can access block, file, and object storage. Many applications that claim to require
block storage will not use it in native block mode, but install a file system on it. Thus they
actually need high-performance storage with “read write once” (RWO) capabilities, which IBM
Storage Fusion can provide out of the box.
Block and file storage are again divided into “container ready storage” and “container native
storage”. Each kind has its own advantages and disadvantages:
“Container ready storage” refers to attaching traditional storage arrays to containers.
This usually limits the scalability to a “scale-up” approach, as the cost of accessing storage in
traditional arrays can vary from pod to pod depending on network and other limitations. Also,
the deployment of traditional storage expansions is usually a large investment that requires
specialized personnel rather than ongoing “maintenance” - type of activities.
“Container native storage” brings the advantages of containerization to the storage

landscape. Capacity is increased via the scale-out approach. Whenever more storage-rich
worker nodes are added to a cluster, each additional node provides not only capacity, but as
well more IO performance and increased caching capabilities leveraging the added CPU,
memory, and network bandwidth.
IBM Storage Fusion HCI System’s global data platform is a “container native storage”. The
HCI is built from 6-20 nodes. Each node can hold up to 10 NVMe drives.
The storage capacity can be increased by either adding NVMe drives to existing nodes or by
adding storage-rich nodes to the HCI cluster. All capacity increases are non-disruptive
operations that are started from the IBM Storage Fusion UI.
Resiliency
Traditional storage arrays provide redundancy by creating a Redundant Array of Independent
Drives RAID across their internal drives. Depending on the RAID level (Raid 5, 6, 10) some
data is used to store the fault tolerance information called parity. But the RAID controller that
is calculating the parity is a single hardware entity that cannot run as distributed software on a
software-defined scale-out storage cluster.
Therefore, most software-defined storage products for containers rely on 2-way or 3-way
replication. This means the complete data is copied and written 2 or 3 times in different failure
zones.
򐂰 2-way replication can only accept one failure and is therefore not suitable for most
production environments.
򐂰 3-way replication can accept two failures but only provides 33% data efficiency.
Dispersed Erasure Coding is a more recent technology to make distributed storage highly
available while providing high data efficiency and short rebuild times.The erasure coding
algorithm slices data into a configurable number of parts, builds the parity information, and
distributes data and parity across the available nodes.
IBM Storage Fusion HCI System uses erasure coding and can be configured to use two
different algorithm variants resulting in different fault tolerances and data efficiencies.

Figure 2-3 shows the IBM Storage Fusion installation dialog which explains the two available
algorithms.
IBM Storage Fusion Install
Figure 2-3 IBM Storage Fusion installation dialog
4+2p = Requires 6 servers, exposes 67% of the raw capacity as usable storage and can
withstand 2 simultaneous failures.
8+3p = Requires 11 servers, exposes 78% of the raw capacity as usable storage and can
withstand 3 simultaneous failures.
This choice must be made with good foresight as it cannot be easily undone. Enterprises will
usually have policies of their own when it comes to error containment vs. efficiency.
2.6 Container Platform Resiliency and High availability

Cloud-native applications can be distributed across multiple clouds and Availability Zones
(AZs). A failure of one cloud platform or one Red Hat OpenShift cluster will not bring down the
whole application due to its built-in redundancy.
Monolithic applications behave differently: They rely on single instances and the availability of
attached persistent Storage, which is usually provided by the hardware level.
In case of a complete node, rack or availability zone failure, workloads can be restarted on the
remaining worker nodes in the Red Hat OpenShift cluster. This process is a lot easier than
restarting a monolithic application in a standby site.
Container platforms like Red Hat OpenShift are resilient by design. Nevertheless, tolerating
the failure of an entire Availability Zone requires careful planning. The Red Hat OpenShift
control plane is built from three master nodes that represent the quorum and tolerate a
maximum network latency of 5ms between nodes.

If three AZs with low latency interconnects are available, the Red Hat OpenShift cluster
including the storage could span all three AZs. See Figure 2-4.
Figure 2-4 Stretched Cluster with three Availability Zones
Each availability zone should host 1/3 of the cluster capacity.
Due to the Red Hat OpenShift network latency requirement of max 5ms, the three AZs would
normally be located in the same campus or building. If a larger physical separation of two AZs
is required, stretching the Red Hat OpenShift cluster is not an option. Two independent Red
Hat OpenShift clusters need to be set up.
IBM Storage Fusion HCI System’s global data platform includes the “Metro-DR” feature,
which provides a synchronous storage replication between two AZs with a maximum network
latency of 40ms which is equivalent to ~150 km signal distance. See Figure 2-5.
Figure 2-5 IBM Storage Fusion HCI System’s global data platform includes Metro-DR
Two IBM Storage Fusion HCI Systems form a Metro-DR cluster with a Tiebreaker in a 3rd
location. The network connection to the Tiebreaker VM in C can tolerate higher network
latencies.
IBM Storage Fusion HCI System with Metro-DR protects monolithic applications that would
otherwise be forbidden in a containerized environment without AZ fault tolerance.

Security
Some security and data management challenges still persist in container environments. A
cyber-resilient storage layer should protect your data in flight and at rest. Keep in mind that
many protocols like nfsv3 are not encrypted. IBM Storage Fusion's global data platform is
based on IBM Storage Scale. It encrypts data at rest and in flight.
More details can be found in the Redbooks IBM Spectrum Scale Security at:
https://www.redbooks.ibm.com/abstracts/redp5426.html
At its core, the Red Hat OpenShift Container Platform relies on the Kubernetes project to
provide the engine for orchestrating containers across many nodes in scalable data centers.
Red Hat OpenShift Container Platform is designed to lock down Kubernetes security and
integrate the platform with a variety of extended components.
See more details at:

https://docs.openshift.com/container-platform/4.11/security/index.html
2.7 Data Protection for Containers

Data from applications in containers require the same level of data protection like applications
in VMs. In times of widespread cybercrime and ransomware attacks, backups should be
protected in an air-gapped environment.
Therefore, data protection is a fundamental part of IBM Storage Fusion. Containers are built
from container images that are stored in a container registry. It is important to have a reliable
backup of the container registry, but there is no need to back up running container images
(again).
What is important though is to back up the information that makes container instances
unique, like passwords, service names or the persistent data of the container. This
information is stored in resources like secrets, services, routes, and a lot more.
Figure 2-6 on page 19 gives an overview of namespace and cluster resources. IBM Storage
Fusion creates a backup of the cluster and namespace resources.

Figure 2-6 Red Hat OpenShift Cluster and Namespace Resources
The persistent data of a container is stored in separate volumes called “persistent volumes
claims” (PVCs) due to the way they are requested from the control plane (as “claim”).
IBM Storage Fusion can take snapshots of PVCs for point-in-time backups. Snapshots are
always stored on the same file system as the primary PVC data. How often these snapshots
are taken and how long they remain available is defined in policies.
In order to protect snapshots from file system corruption, malicious destruction or physical
damage, they can be periodically copied to an Object Storage service or to an external
backup server. Backup and restore are managed from within the IBM Storage Fusion user
interface. See Figure 2-7.
Figure 2-7 Backup and restore are managed within the IBM Storage Fusion user interface

2.8 Red Hat OpenShift made easy: Day2 Operations with IBM
Storage Fusion
This section describes how Red Hat OpenShift is made easy with Day2 Operations with IBM
Storage Fusion.
Hardware monitoring
Today's servers and switches have built-in hardware redundancy like multiple fans and power
supplies. But the best redundancy does not help if failed components are not identified and
replaced as soon as possible, minimizing the risk of double failures.
Red Hat OpenShift does not include hardware monitoring of the underlaying infrastructure
out-of-the-box. In particular, when Red Hat OpenShift runs on top of a hypervisor, hardware
monitoring needs to be implemented on the hypervisor layer or below. If Red Hat OpenShift is
deployed on bare metal, infrastructure monitoring and alerting has to be implemented by the
user via 3rd party tools.
IBM Storage Fusion HCI System includes a full stack hardware monitoring and call home
functionality. Events are raised and identified failed components can be replaced ASAP. This
functionality is only available with the HCI version, not with IBM Storage Fusion SDS.
Single Pane of Glass UI

IBM Storage Fusion HCI System features a “single pane of glass” user interface.
The Infrastructure Dashboard provides an overview of the complete hardware and software
stack (see Figure 2-8). This holistic view eases troubleshooting and allows diving deeper into
each component with a single click.
IBM Storage Fusion SDS includes a UI with the same look and feel but without the
infrastructure dashboard.
IBM Storage Fusion
Figure 2-8 IBM Storage Fusion UI Dashboard
The UI also contains an “Applications” section in which details about all Red Hat OpenShift
projects can be found. Backup policies can be easily assigned to applications and an
overview of the storage consumption and status is provided.

Infrastructure as Code
IBM Storage Fusion HCI System does not include any management host or console.
As illustrated in the architecture diagram shown in Figure 2-9, all software components are
fully container-native and run on top of Red Hat OpenShift.
Figure 2-9 IBM Storage Fusion HCI System Software Architecture Diagram
The IBM Storage Fusion software stack is deployed as an Red Hat OpenShift operator and
updated via the Red Hat OpenShift operator lifecycle management concept (OLM).
IBM Storage Fusion adds many custom resource definitions (CRDs) to the Red Hat
OpenShift cluster. This enables IBM Storage Fusion to store its configuration and current
states in custom resources (CRs) in the Red Hat OpenShift cluster configuration database
(etcd).
This architecture provides many advantages for “Operations”:
Changing configurations by editing human-readable YAML files is the Red Hat

OpenShift-native way of managing the cluster. IBM Storage Fusion custom resources can be
accessed and changed either from the Red Hat OpenShift UI or from the command line
interface CLI. This eases the automation of configuration changes like creating backup
policies or adding backup policies to namespaces.
The IBM Storage Fusion HCI System software includes compute and network operators that
are not available for custom-built Red Hat OpenShift clusters on bare metal or a hypervisor.
These operators take care of the hardware management within the IBM Storage Fusion HCI
System and transmit details like firmware levels, temperatures, hardware component failures,
network vlans, packet counters per network port, and more to Red Hat OpenShift.
This “infrastructure as code” can be used to ease troubleshooting and allow a DevOps
operating model.

Updating the Complete Red Hat OpenShift Software Stack
In order to harden the system against cybercrime and to be compliant with various security
audits, updating only Red Hat OpenShift is not sufficient.
The full software stack including the firmware of servers and switches, the hypervisor
software and CSI drivers for storage systems need to be updated as well. See Figure 2-10.
Before the stack is updated, the interoperability of the new versions of the components need
to be verified.
For each IBM Storage Fusion release, IBM delivers a supported and pre-tested software
bundle including:
򐂰 IBM Storage Fusion operator
򐂰 Data Protection software
򐂰 Global Data Platform (SDS)
򐂰 Firmware updates for servers and switches (HCI only)
Figure 2-10 IBM Storage Fusion Bare metal Red Hat OpenShift Stack
While Red Hat OpenShift and the IBM Storage Fusion operator are updated from the
underlying Red Hat OpenShift UI, all other components are updated from the IBM Storage
Fusion UI.
A custom-built Red Hat OpenShift stack on top of a Hypervisor includes more layers from
different vendors. The software needs to be obtained from many sources, each coming with
its own installation instructions and interoperability guidelines.

The numerous teams (Figure 2-11) involved in larger enterprises increase the complexity and
coordination effort of maintenance activities.
Figure 2-11 Custom-built Red Hat OpenShift stack on top of a Hypervisor includes more layers from
different vendors adding complexity

3
Chapter 3. Flavors of IBM Storage Fusion

This chapter describes flavors of IBM Storage Fusion implementations, major differences
between HCI (hardware) and SDS (software) variants, and which choices to make to
right-size the product for current needs while retaining the ability to scale out at a future stage.

3.1 Integration with legacy VM infrastructures
IBM Storage Fusion goes one step further in also allowing the integration of legacy app
implemented as VMs. This aspect is important for hybrid applications, where some parts like
a legacy database, cannot yet be containerized, however these type of scenarios need to
obey the same deployment, monitoring, availability, and security paradigms.
Remember: VMs are still machines, even though they are virtualized from the underlying
hardware. Therefore, they need to be managed as such. In contrast, container platforms do
not consider machines, but rather services, templates, recipes, and resource claims. These
resources can include instances of virtual machines.
The more we modernize our applications, the more we will be able to write “infrastructure as
code”; namely, recipes and services/resource claims. The classic concept of a virtual
machine that conceals all mutual dependencies and security aspects is not adequate for
modern “agile” or “cloud-native” programming and dev-ops-style deployment.
3.2 Three Tier Architecture System

If we look at the traditional way an IT system is built, we have three independent tiered layers
known as the 3 tier “siloed” architecture. These are as follows:
򐂰 Network Tier
򐂰 Server Tier
򐂰 Storage Tier
Figure 3-1 shows the typical construction of a three tier architecture system.
Figure 3-1 Three tier architecture system

In this example we have separate tiers where the storage, server and network elements are
all installed and run as individual items. This could be from potentially different vendors and
thus separate interfaces or controls for each one. The complexity, operation and running of
the complete system thus requires multiple users’ interfaces and training for the operation
teams. A hyper-converged infrastructure attempts to address these issues, and by doing this,
reduces the complexity and costs of running all the separate elements of the three tiered
“siloed” approach.
3.3 Hyper-Converged Infrastructure (HCI)

The HCI methodology combines servers, storage, and network plus the required software
defined storage components to complete the IT system in its entirety. Instead of the traditional
separation of compute, network, and storage resources, HCI employs a simplified and unified
way to couple all the resources together and have a single management operational interface.
HCI also provides a way to give easily adapted scalability functions in the event that the
system needs to grow to accommodate expanding and more demanding workload.
Figure 3-2 shows the layout of a typical HCI system.
Figure 3-2 HCI system overview
Currently one of the most common adoptions of HCI technology is Virtual Desktop
Infrastructure (VDI) application. IBM’s vision of HCI is beyond the simplification of the
management and legacy VDI workloads. In the modern HCI, it should optimize for container
deployment for hybrid cloud and Edge environments. It should support enterprise features
such as seamless backup and restore, disaster recovery, enterprise security and high
availability. The modern HCI should not be an infrastructure silo. It should be able to create a
unified storage platform to allow seamless application mobility and data sharing from clouds
to the edge.
Chapter 3. Flavors of IBM Storage Fusion 27

3.4 IBM Storage Fusion Options
IBM Storage Fusion is offered either with a bare metal RHOCP deployment in an x86-based
Hyper Converged Infrastructure (HCI) appliance form factor, or as a pure Software-Defined
Storage (SDS) solution.
IBM Storage Fusion is built from proven, enterprise-hardened technologies like IBM Storage
Scale and IBM Spectrum Discover (now an integrated IBM Storage Fusion service called
Data Cataloging) and provides a bare metal foundation for the rapid deployment and easy
management of any RHOCP environment. We've brought the entirety of our rich heritage and
vast IP portfolio in supercomputing, AI/ML, and industrial-strength data resilience to bear on
the design of Fusion. The result is a performance- and cost-optimized platform that unlocks
the full potential of RHOCP - at any scale.
3.4.1 IBM Storage Fusion Software Overview

The IBM Storage Fusion software consists of several building blocks which together provide
the function and features of the software solution.
Figure 3-3 shows the IBM Storage Fusion software layout overview.
Figure 3-3 IBM Storage Fusion software layout
IBM Storage Fusion and the inherent multi-cloud deployment model of IBM Storage Fusion
Data Foundation, in conjunction with Red Hat OpenShift itself, are key to addressing and
enabling the agility, simplicity, and productivity improvements that create efficiencies for
customers.

IBM Storage Fusion Data Foundation exposes physical storage to the applications that are
deployed on Red Hat OpenShift. Kubernetes abstracts the storage using container storage
interfaces (CSI) and Storage Classes, but it does nothing more to help manage the storage
nor to provide uniformity of presentation and usage across multiple deployment models. IBM
Storage Fusion Data Foundation delivers consistent volume types, resilience, and tiering
wherever the storage layer is provisioned.
The IBM Storage Fusion Data Foundation documentation is split into sections for each type of
environment describing how to deploy available storage in each one. At the edge, there is a
new option for very small deployments of Single Node Red Hat OpenShift (SNO) matched by
IBM Storage Fusion Data Foundation with a more specialized deployment of Local Volume
Manager (LVM) instead of Ceph, and MultiCloud Object Gateway.
In public cloud environments, there are managed service offerings for storage on Red Hat
OpenShift Service on AWS (ROSA) and Red Hat OpenShift Dedicated. In all environment
types, there are self-managed options in which the client can choose their preferred
infrastructure and storage model.
3.4.2 IBM Storage Fusion HCI System Appliance Overview

IBM’s model for the complete HCI appliance is to combine all the hardware outlined above
together with the software components listed below, plus a custom-built management user
interface (UI), all running as containerized applications within the IBM Storage Fusion HCI
System appliance.
This gives the IBM Storage Fusion HCI System the following characteristics:
򐂰 Integrated HCI appliance for both containers and VMs using Red Hat OpenShift
򐂰 Highly scalable containerized file system with erasure using IBM Storage Scale ECE
򐂰 Data resilience for local and remote backup and recovery using IBM Storage Protect Plus
򐂰 Simple installation and maintenance of hardware and software through the custom-built UI
򐂰 Global data platform stretching from public clouds to any on-prem or edge locations
򐂰 IBM Cloud Satellite® and Red Hat ACM (Active Cluster Management) native integration
򐂰 Ready for AI applications with optional GPUs and global data for better AI
򐂰 Starts small with 6 servers and scales up to 20 (with HPC GPU enhanced options).
So, in summary, these are the highlights of the IBM Storage Fusion HCI System appliance
and all the advantages of this hyper-converged approach using the outlined building blocks.

3.4.3 IBM Storage Fusion HCI System Software Overview
The IBM Storage Fusion HCI System software consists of several building blocks which
together provide the function and features of the software solution within the HCI appliance.
Figure 3-4 shows the IBM Storage Fusion HCI System software overview.
Figure 3-4 IBM Storage Fusion HCI System software
Some of the main points of the HCI software stack are:

򐂰 Red Hat OpenShift on Bare Metal is installed across all nodes with the bare metal OS
being Red Hat CoreOS (RHCOS).
򐂰 Next, a set of HCI management services are installed in blue on top of Red Hat OpenShift
that manages the compute, storage and networking:
򐂰 The HCI UI operator provides a single pane of glass to manage IBM Storage Fusion and is
integrated with Single Sign On provided by Red Hat OpenShift.
򐂰 The storage operator deploys a best practice installation of IBM Storage Scale ECE.
򐂰 The compute operator manages the physical servers including firmware upgrade, power
on/off, and scaling the number of nodes in the cluster.
򐂰 The networking operator manages the switches including switch firmware and
configurations.
򐂰 The HCI operators monitor the health of the system for problem resolution, provide
performance monitoring capabilities, present capacity utilization.
– Allows the user to know when you need to add more drives for storage capacity or
more physical servers for more compute capacity.
– Events are natively integrated with Red Hat OpenShift and can Call Home to IBM in the
event of a physical infrastructure or software problem with automatic log collection and
upload.

򐂰 IBM Storage Protect Plus is integrated with the backup operator and can backup both the
IBM Storage Fusion control plane as well as deployed client applications on Red Hat
OpenShift.
򐂰 Finally, the upgrade operator orchestrates the upgrade of firmware on the hardware, Red
Hat OpenShift, the HCI Management Stack, IBM Storage Scale ECE and IBM Storage
Protect Plus.
3.4.4 IBM Storage Fusion HCI System Hardware Overview

The IBM Storage Fusion HCI System consists of several hardware building blocks which
together provide the function and features of the appliance.
Figure 3-5 shows the HCI hardware overview.
Figure 3-5 HCI hardware overview
The base configuration includes:

򐂰 IBM 9155 Model R42 42U rack including the power distribution units (PDU’s)
򐂰 IBM 9155 Model S01 - 2x Ethernet high-speed switches
򐂰 IBM 9155 Model S02 - 2x Ethernet management switches
򐂰 IBM 9155 Model C01 - 6x 32 Core Storage/Compute servers with 2 NVMe drives/servers
– OR
򐂰 IBM 9155 Model C05 - 6x 64 Core Storage/Compute servers with 2 NVMe drives/servers
– The base system will always include the following models (quantity per model)
• 6 x C01 or 6 x C05 or 3 x C01 + 3 x C05
• 2 x S01 Ethernet high-speed switches
• 2 x S02 Ethernet management switches

• 1 x R42 Rack
• 1 x TF5 KVM Feature (optional)
– The base system can be extended with the following models:
• C00 - 32 Core Storage/Compute servers with no drives
• C01 - 32 Core Storage/Compute servers with 2 NVMe drives/servers
• C04 - 64 Core Storage/Compute servers with no drives
• C05 - 64 Core Storage/Compute servers with 2 NVMe drives/servers
• F01 - 2x 16 Core servers used for Active File Management
• G01- 2x 24 core server with 3x NVIDIA GPUs.
– The C00, C01, C04, and C05 models of servers can be intermixed in the system, but
the base 6 servers must have NVMe drives installed, so currently have to be C01 or
C05 types.
Notes: The server in rack position seven (RU7) is the provisioner node, connected to
the optional KVM feature. On older systems, this KVM was a standard feature, however
on current system orders, the KVM feature is optional.
This KVM is used by the IBM SSR to configure the initial systems with network and
base Red Hat OS to then move to the OCP and Fusion software install. The servers in
positions two, three and four (RU2, RU3, and RU4) become the Red Hat OpenShift
control plane.
On systems without the KVM feature, the SSR will use a service laptop to connect into
the node in rack position seven (RU7) and then perform the initial install phase.
IBM Storage Fusion HCI System 2.5.2 also offers the following new enhancements:
򐂰 Install IBM Storage Fusion HCI System components in your own rack
– IBM Storage Fusion HCI System components can be ordered for installation into
client-supplied racks. When ordering, clients have the option to order an IBM Storage
Fusion HCI System fully racked from the IBM factory or as components that can be
installed into a client-provided rack.
– For more information see here: Installing the HCI System
򐂰 High Availability Cluster Support
– IBM Storage Fusion HCI Systems can be configured into a three-rack high availability
(HA) cluster. This configuration is commonly referred to as a 3 availability zone (3AZ)
deployment, a strategy useful for ensuring application availability.
– You can connect three IBM Storage Fusion HCI System racks to create a single large
Red Hat OpenShift Container Platform and IBM Spectrum Scale ECE cluster.
– The components in each of the three racks are the same as a single rack.
– For more information see here: High Availability Cluster
򐂰 Expansion Rack Support
– IBM Storage Fusion HCI System can add new IBM racks to expand your Red Hat
OpenShift Container Platform cluster.
– For more information see here: Adding Racks

򐂰 Tailored Power and Network Options
– IBM Storage Fusion HCI Systems will only include the number of power distribution
units (PDUs) and network cables required to support the configured order. MES orders
will add PDUs, and cables as needed.
– For more information see here: Power Configuration Calculations
More detailed information is available here where the new full feature set for the IBM Storage
Fusion HCI system is explained.
3.4.5 Scalability in HCI

IBM Storage Fusion HCI System is a highly scalable appliance, not only in computing power,
but also in internal storage capacity within one frame. The IBM Storage Fusion HCI System is
also scalable buy adding extra nodes (either storage/compute or just compute only) for
increased power and storage.
Server / Disk upgrade options

The server and disk (SSD) upgrade options are as follows:
򐂰 Increased storage by adding pairs of drives to storage/compute servers
– 7.68TB NVMe PCIe Gen4 drives/server to a max of 10 drives/server
򐂰 Increased compute power / storage by adding servers
– Sold in pairs
򐂰 AFM (Active File Manager) delivered as a pair of servers
򐂰 A pair of 2U GPU servers, each with 3x NVIDIA A100 GPUs
Server Memory Options

The following memory upgrade options per server are available:
򐂰 Standard memory in the 9155-C01 and 9155-C00 servers is 256GB
– Memory can now be upgraded in these C01 and C00 servers
– Note: AFM and GPU servers are not eligible for memory upgrades
– Memory can be upgraded in manufacturing
– The manufacturing update increases the RAM to 512GB
– All C01 and C00 servers must be ordered with the same memory configuration
򐂰 Upgrade in the field (MES)
򐂰 Upgrade C00 or C01 servers to either 512GB or 1024GB
– The 1024GB upgrade cannot be done in manufacturing
Note: Because of power requirements, only systems with 3-phase power or 60A
single-phase power are eligible for the 1024GB upgrade. Systems using 30A
single-phase power cannot support this memory upgrade.
򐂰 Standard memory in the 9155-C04 and C05 server is 1024GB

– Memory can now be upgraded in these servers
– Memory can be upgraded in manufacturing

– The manufacturing update increases the RAM to 2048GB
– All C04 and C05 servers must be ordered with the same memory configuration
򐂰 Upgrade in the field (MES)
– Upgrade C04 and C05 servers to 2048GB
3.4.6 Configuring the IBM Storage Fusion HCI System with IBM Storage
Modeller
The IBM Storage Fusion HCI System appliance is configured using a tool call IBM Storage
Modeller (StorM). IBM Storage Modeller is an IBM sales process tool that is designed to
model storage solutions for customer applications. This tool is for use by IBM and Business
Partner technical sellers.
Storage Modeller is developed by IBM and functionality is being delivered over time. Storage
Modeller is designed to integrate with upstream and downstream sales tools, with the focus
on optimizing seller workflow and reducing time from Opportunity ID to actual sale of a
solution. Storage Modeller is a cloud-based tool accessed through a web browser-based user
interface.
Figure 3-6 shows the Storage Modeller process for integration with IBM sales tools.
Figure 3-6 Typical storage system design in the StorM tool
As the StorM tool is used, and the configuration created, the UI display will show a graphical
build of the system. More information on the StorM tool and a virtual tour can be found here:
IBM Storage Modeller
Figure 3-7 on page 35 shows the IBM StorM graphical output representation of the HCI
appliance configuration.

Figure 3-7 IBM StorM graphical output
The left hand section of the UI allows the user to select the number and type of servers, and
storage per server, both the system memory and the number of NVMe drives (minimum of
two per server to a maximum of ten per server).
Note: The number of NVMe drives must be symmetrical across all of the six base servers
and any additional servers added. The StorM tool will check for this rule.
The right hand section of the UI shows the license information, raw vs. usable storage and the
selected hardware resources chosen with the relevant IBM feature codes.
3.4.7 VMware versus bare metal

IBM Storage Fusion HCI System runs Red Hat OpenShift on bare metal servers as part of the
appliance delivery model. This ensures that the overhead on the VMware hypervisor costs
are eliminated.
For more information on the details of each type of installation available, refer to the IBM
Document pages for IBM Storage Fusion here:
IBM Documentation for Deploying IBM Storage Fusion

Unlike IBM Storage Fusion HCI System, the IBM Storage Fusion Software solution does not
come with any hardware, so the customer can choose the x 86 hardware platform they want
to run on. IBM Storage Fusion Software currently runs VMware hypervisor only. Future plans
include running on bare metal and a variety of cloud provider platforms, including of course
IBM Cloud.
Figure 3-8 shows the variations on the IBM Storage Fusion HCI System and Software only
versions and which platform each will run on.
Figure 3-8 IBM Storage Fusion flavors
The components within each IBM Storage Fusion offering are as follows:
IBM Storage Fusion HCI

IBM Supplied hardware as follows:
򐂰 Minimum of six compute servers with internal storage,
򐂰 Two switches for high-speed data flow
򐂰 Two switches for management control
򐂰 Six power distribution units (PDU)
򐂰 IBM 42U approved rack
IBM Storage Fusion Software

The IBM Spectrum Fusion software only version at 2.3 and 2.4, and IBM Storage Fusion
software 2.5 and above, have the ability to run on the following:
򐂰 VMware hypervisor as a guest
򐂰 Run on base metal via these hardware providers:
– Lenovo
– Cisco
– Dell
– Hewlett Packard (HPE)
– Other commodity hardware

The IBM Spectrum Fusion software only version at 2.3 and 2.4, and IBM Storage Fusion
software 2.5 and above, have the ability to run on the following:
򐂰 IBM Cloud
򐂰 AWS - Amazon Web Services
򐂰 IBM zLinux
3.4.8 Use cases for IBM Storage Fusion software

The main theme on IBM Storage Fusion software is for customers who need to run mission
critical applications providing data services storage, with high availability (HA) and disaster
recovery (DR) plus full backup and restore option. IBM Storage Fusion software can do all of
this with a minimum of install time and can provide all these functions.
Also some other considerations are as follows:

򐂰 IBM Spectrum Storage for Cloud Paks (SS4CP) is no longer being offered, so the IBM
Storage Fusion is now the preferred option
– Customers who originally acquired SS4CP are now moving to IBM Storage Fusion
software as an ongoing solution
򐂰 Customers who have purchased IBM Storage Scale and want to use it with OCP but
CNSA+CSI is too complicated to deploy and operate, plus they also seek a method for full
Backup and Recovery integration
򐂰 Official DR support and global data sharing will most likely also only come with IBM
Storage Fusion
򐂰 Customer is expanding into Public Cloud and only the IBM Storage Fusion software
version will bring official support there
򐂰 Client is building Development/Test/Pre-prod clusters along side a production level The
IBM Storage Fusion HCI System
So in summary IBM Storage Fusion software allows you to solve the same problems BUT we
meet you on your platform of choice.
One very common client scenario we see today is IBM Storage Fusion Data Foundation +
Cloud Pak for Data and in particular IBM Storage Fusion providing the backup/restore
capability for a complex Cloud Pak installation.

4
Chapter 4. IBM Storage Fusion Terminology

This chapter describes the terminology used for containers environments and more
specifically for IBM Storage Fusion. It also covers some more advanced functions included in
the product.
There is also a section on what is included with the various releases of the IBM Storage
Fusion software.

4.1 Operators
An Operator is a method of packaging, deploying, and managing a container-native
application. A container-native application is an application that is both deployed on
Kubernetes and managed using the Kubernetes APIs and tooling.
Note: Kubernetes (also known as k8s or “kube”) is an open-source container orchestration

platform that automates many of the manual processes involved in deploying, managing,
and scaling containerized applications.
Operators are pieces of software that ease the operational complexity of running another
piece of software. They act like an extension of the software vendor's engineering team,
watching over a Kubernetes environment (such as Red Hat OpenShift Container Platform)
and using its current state to make decisions in real time.
These Operators allows for simple management of the IBM Storage Fusion HCI System
appliance and allows IBM to build the appliance management software using the latest
software development practices including CI/CD pipelines and online software delivery via
entitled registries for simplified upgrades. IBM Storage Fusion HCI System also supports a
fully air-gaped or disconnected installation.
4.2 Upgrading
IBM Storage Fusion HCI System incorporates an upgrade operator which orchestrates the
upgrade process of firmware on the hardware, and application software of Red Hat
OpenShift, the Appliance Management Stack, IBM Storage Scale ECE and IBM Storage
Protect Plus.
4.3 Management UI
IBM Storage Fusion provides a management or appliance UI for managing the HCI System
appliance. This UI has the ability to monitor and control every area of the appliance. The
appliance UI operator provides a single pane of glass to manage IBM Storage Fusion HCI
System and is integrated with Single Sign On provided by Red Hat OpenShift. The appliance
UI operators monitor the health of the system for problem resolution, provide performance
monitoring capabilities, present capacity utilization so that as a user you know when you need
to add more drives for storage capacity or more physical servers for more compute capacity.
4.4 IBM Storage Scale Erasure Code Edition (ECE)

By default, when IBM Storage Fusion HCI System is installed, a best practice installation of
IBM Storage Erasure Code Edition (ECE) is installed so that software deployed on Red Hat
OpenShift can easily consume the storage using dynamic provisioning.

4.5 Red Hat OpenShift Virtualization
With Red Hat OpenShift Virtualization comes the ability to deploy virtual machines (VMs) side
by side with containers using the same Red Hat OpenShift control plane.
4.6 IBM Storage Fusion Data Cataloging service

The Data Cataloging service is modern metadata management software that provides data
insight for exabyte-scale heterogeneous file, object, backup, and archive storage on premises
and in the cloud. It can help you manage your unstructured data by reducing data storage
costs, uncovering hidden data value, and reducing the risk of massive data stores.
Note: IBM Spectrum Discover functionality is now available as an integrated service within
IBM Storage Fusion called Data Cataloging.
If you want to use Data Cataloging, you must take into consideration the following system
requirements: IBM Storage Fusion system requirements
After the system requirements for Data Cataloging are met, you can install the Data
Cataloging service.
Note: Data Cataloging is applicable only for IBM Storage Fusion and not for IBM Storage
Fusion HCI System. It is enabled starting with IBM Storage Fusion 2.5.1 software.
After Data Cataloging is installed you will need to check IBM Storage Software requirements
and then configure the data source connections to meet your needs such as the following
examples:
IBM Storage Scale data source connections
You can create an IBM Storage Scale data source connection, scan a data source,
and manually initiate a scan.
IBM Storage Archive data source connections
You can define tags and policies in Data Cataloging based on values that are
derived from IBM Storage Archive metadata to help in searching and categorizing
files.
IBM Cloud Object Storage data source connection
You can create the IBM Cloud Object Storage (COS) connection and initiate a
scan.
Data Cataloging and S3 object storage data source connections
Use this information to understand how Data Cataloging works with S3-compliant
object store.
Scanning an Elastic Storage Server data source connection
Use the Data Cataloging GUI to scan an Elastic Storage Server (ESS) data
connection.
Chapter 4. IBM Storage Fusion Terminology 41

IBM Storage Fusion documentation for Data Cataloging will show you how to use the
functions to manage your unstructured data by managing metadata and content search
policies, managing tags, using managing applications, using the IBM Spectrum Discover
applications catalog, reporting, and many more functions.
4.7 IBM Cloud Satellite

IBM Cloud Satellite extends the capabilities of IBM Cloud into virtually any location including
the datacenter or at the edge. It comes with an integrated UI that can deploy IBM Cloud
Satellite hosts in a matter of minutes. As your business needs grow, IBM Storage Fusion HCI
System makes it easy to expand capacity from both a compute (CPU and memory) and
storage perspective to accommodate more IBM Cloud Satellite workloads.
Once the IBM Storage Fusion HCI System is configured as an IBM Cloud Satellite location,
from the IBM Cloud console you can then deploy IBM Cloud Services such as Red Hat
OpenShift clusters and other IBM Cloud Satellite enabled services. IBM Storage Fusion HCI
System in this case serves as a turnkey IBM Cloud Satellite location with integrated
infrastructure, capacity management, system monitoring, call home, upgrades, and more.
4.8 Red Hat Advanced Cluster Management (RHACM)

Red Hat Advanced Cluster Management (RHACM) can manage multiple Red Hat OpenShift
or other Kubernetes clusters and can manage the resources running on those clusters from a
single pane of glass. Since IBM Storage Fusion HCI System is based on Red Hat OpenShift,
RHACM can natively manage multiple IBM Storage Fusion HCI System systems. RHACM
can run anywhere, including on an IBM Storage Fusion HCI System.
With RHACM you can:

򐂰 Manage Red Hat OpenShift and Kubernetes clusters from a single pane of glass.
򐂰 Mange the Kubernetes resources on these clusters and easily make changes to multiple
resources on multiple clusters.
򐂰 Deploy applications to and across Red Hat OpenShift clusters including move an
application from one cluster to another using placement policies.
򐂰 Create policies that enforce the configuration of Kubernetes resources across multiple
Red Hat OpenShift clusters.
Note: Both IBM Cloud Satellite and Red Hat Advanced Cluster Management can be
used together to build a hybrid cloud solution with single pane of glass management.
IBM Cloud Satellite is used to deploy IBM Cloud services to on-prem and edge
locations with IBM Storage Fusion HCI System. Red Hat Advanced Cluster
Management is used to manage hybrid cloud resources and applications across
multiple IBM Storage Fusion HCI Systems, public clouds, and more.

4.9 IBM Storage Protect Plus
IBM Storage Fusion HCI System comes integrated with IBM Storage Protect Plus. IBM
Storage Protect Plus provides backups of the IBM Storage Fusion HCI System control plane,
as well as backups of running applications on Red Hat OpenShift.
IBM Storage Protect Plus can take snapshots of applications and save them locally or save
them off the rack to an external backup location, such as an IBM Storage Protect Plus vSnap
server, that is connected to on-prem or public cloud storage.
4.10 Call Home for all IBM Storage Fusion HCI System
IBM Storage Fusion HCI System has a fully integrated Call Home function which covers both
the IBM Storage Fusion HCI System hardware infrastructure and software with automatic log
collection and upload for rapid problem resolution. The call home function will take any
information passed to it from the appliance UI that is deemed to be critical enough to require
further analysis by IBM Support.
This call home function will connect securely to IBM and raise a CSP (Cognitive Support
Platform) ticket with IBM support. This CSP ticket will then be passed to the relevant IBM
support specialist to analyze and progress the call. Events are natively integrated with Red
Hat OpenShift and can call home to IBM in the event of a physical infrastructure or software
problem with automatic log collection and upload.
4.11 Active File Management (AFM)

Active file management (AFM) is a scalable, high-performance, file system caching layer, and
enables sharing of data across clusters, even if the networks are unreliable or have high
latency.
IBM Storage Fusion HCI System supports Active File Management (AFM) through the use of
a pair of specialized AFM servers, ordered as a feature code.
The competitive differentiator is not just the fact that IBM has fused a storage platform with
storage services but with the storage platform we have leveraged years of investment in a
global parallel file system with global data access and advanced file management (AFM).
Applications simply access a local file or directory but are accessing data from information
that is stored in a public cloud bucket or legacy data from a remote system 1000s of miles
away. Some of the benefits of AFM are listed here:
򐂰 You can use AFM to create associations between IBM Storage Fusion clusters.
򐂰 With AFM, you can implement a single namespace view across sites around the world by
making your global namespace truly global.
򐂰 By using AFM, you can build a common namespace across locations, and automate the
flow of file data. You can duplicate data for disaster recovery purposes without suffering
from wide area network (WAN) latencies.
򐂰 Individual files in the AFM filesets can be compressed. Compressing files saves disk
space.
򐂰 Namespace replication with AFM occurs asynchronously so that applications can operate
continuously on an AFM fileset without network bandwidth constraints.

The active file management (AFM) feature of IBM Storage Fusion continues to expand its
data protection configuration options for disaster recovery (DR) scenarios by introducing the
IBM Storage Scale feature AFM-based disaster recovery (AFM async DR).
򐂰 AFM async DR uses asynchronous data replication between two sites. The replication
mechanism is based on the AFM file set level replication DR capability with a strict
one-to-one active-passive model.
򐂰 In this model, the sites are represented as being primary and secondary, and the
replicated file sets are mapped one-to-one.
򐂰 The active file management masks wide area network (WAN) latencies and outages by
using IBM Storage Scale to cache massive data sets and provide asynchronous data
movement between cache and home.
򐂰 The AFM async DR feature is focused on accomplishing business recovery objectives and
targeting disaster recovery scenarios by replicating all data asynchronously from a
primary site to a secondary site.
򐂰 From a disaster recovery (DR)/ high availability (HA) perspective, IBM supports complete
file system, file set, and file data protection capabilities. Both synchronous and async
replication are supported. Multi-zones are supported.
򐂰 Snapshots for end-to-end and provide a fast, easy way to protect data locally from
corruption, modification, or deletion. Additionally, IBM Storage Fusion provides direct
access to application checksum in ESS or IBM Storage Scale ECE.
All these capabilities put together provide not just resilient data but also location aware and
location independent access to data across geographic distance, multi-cluster distance, and
multi-availability zones. Distance is very important and is a requirement for a lot of enterprise
workloads. Consequently, data resiliency has become increasingly important as cyber
resiliency is part of board level discussions.
4.12 AI Support (AI GPU)

Artificial intelligence (AI) and machine learning (ML) require considerable elasticity and
dispersal of compute and storage requirements to gather data, present it to the data scientists
who refine data and build models for training, as well as production purposes.
Furthermore, there are lifecycles in which those models are assessed and improved on a
continuous basis as follows:.
򐂰 The training datasets and production models all need easy to manage storage
provisioning to support the pipelines that are now prevalent in both DevOps and AI
continuous delivery environments.
򐂰 Convergence of applications and infrastructure is threefold; computing power for AI/ML
with graphic processing unit (GPU) assist, and co-location of data for training and high
performing production models.
IBM Storage Fusion HCI System can be used to optimize AI workflows with modern container
applications. It does this through the use of a pair of specialized GPU servers ordered as a
feature code.
If a customer is using NVIDIA for AI IBM Storage Scale it can preprocess or process at the
edge with NVIDIA A100 GPUs and NVIDIA Red Hat Operator and tools and then
transparently leverage that data in the cloud or directly on another larger more compute
intensive NVIDIA DGX A100 system with the IBM ESS 3000 or ESS 3200 high performance
storage system.

Advantages of this are:
򐂰 Accelerate AI/ML workloads with NVIDIA A100 GPU Nodes
򐂰 Process data generated at the edge, flow ‘the interesting bits’ to a central location for
further processing
򐂰 Train ML models from data in IBM Storage Fusion HCI System’s high-performance NVMe
Flash storage
So, in summary the optional pair of GPU accelerated nodes for AI workloads that leverage the
latest NVIDIA A100 GPU, can also leverage the latest Red Hat OpenShift Operator for AI
from NVIDIA.
This will make creating and running AI workloads on IBM Storage Fusion HCI System even
easier and shows how this can be integrated into an overall AI enhanced workflow.
4.13 Synchronous Storage Replication with Metro-DR

You can have a stretched IBM Storage Scale cluster by using which the file system is made
available in two availability zones of a data center or across multiple data centers within a
metro area.
These two independent clusters run active-active so the container applications can be moved
in case of failure. As an Red Hat OpenShift admin, you can deploy IBM Storage Fusion HCI
System in data center site 1 and data center site 2, resulting in Red Hat OpenShift Container
Platform clusters that run at each site. Two clusters are configured such that they share a
synchronously replicated storage layer, allowing for data mirroring with zero Recovery Point
Objective (RPO) between the two clusters.
IBM Storage Fusion ensures data availability between the two clusters (both the data and the
PV/PVCs). However, you are responsible for managing application deployments between the
cluster. Here, both Red Hat OpenShift Container Platform clusters host active workloads at
the same time, rather than requiring one Red Hat OpenShift Container Platform cluster to be
a cold standby.
The Metro-DR setup includes two sets of networks, namely internal storage network and
Red Hat OpenShift network. In Metro-DR, the internal storage network of one site needs to
connect to the other site and tiebreaker, that is, externalize your storage network. Similarly,
the Red Hat OpenShift network of one site needs to connect to the other site.

Figure 4-1 shows the connectivity between Site 1 and Site 2.
Figure 4-1 Connectivity between Site 1 and Site 2
For more info see here: Metro-DR
4.14 IBM Storage Fusion Reference Information and Links

The IBM Documentation Center has a listing of which features, and changes are applicable to
each level of IBM Storage Fusion software release. For the most recent release check here.
Here are the details of each release and links to the IBM Documentation Center for more
detailed information.
4.14.1 New features and changes in 2.5.2

Listed here are some of the new features in IBM Storage Fusion 2.5.2
򐂰 High Availability Cluster Support
– IBM Storage Fusion HCI System lets you connect three IBM Storage Fusion HCI
System racks together to create Red Hat OpenShift Container Platform clusters.
򐂰 Expansion Rack Support
– IBM Storage Fusion HCI System can add new IBM racks to expand your Red Hat
OpenShift Container Platform cluster.
򐂰 IBM Storage Fusion HCI Components in Your Own Rack
– Install IBM Storage Fusion HCI System components in your own rack
– IBM Storage Fusion HCI System can be purchased without an IBM rack and instead is
installed in a rack that you provide in your data center.

򐂰 New supported component versions in IBM Storage Fusion 2.5.2
– Red Hat OpenShift Container Platform 4.12 for IBM Storage Fusion and Red Hat
OpenShift Container Platform 4.10 for IBM Storage Fusion HCI System
– IBM Storage Fusion Data Foundation 4.12
– IBM Spectrum Scale 5.1.7.0
– IBM Spectrum Scale Container Native 5.1.7.0 for IBM Storage Fusion HCI
– Data Cataloging 2.1.0
– Backup & Restore 2.5.2
– For supported IBM Cloud Paks versions, see IBM Cloud Paks support for IBM Storage
Fusion
Note: The 2.5.1 release is only for IBM Storage Fusion and not for IBM Storage Fusion
HCI System. The IBM Storage Fusion HCI System features are picked up in release 2.5.2.
Listed here are some of the new features in IBM Storage Fusion 2.5.1.
򐂰 Dynamic storage provisioning using IBM Storage Fusion Data Foundation
– You can configure Dynamic storage device from the IBM Storage Fusion UI for IBM
Cloud and VMware platforms. Support is also available to configure IBM Storage
Fusion Data Foundation encryption from the IBM Storage Fusion user interface.
򐂰 New platform support for Global Data Platform service
– Global Data Platform service is now supported on ROSA platform. You can connect
and add remote file system from the IBM Storage Fusion user interface.
򐂰 Policy frequency can be realized in units of minutes or hours with the new service
– Multiple policies can be assigned to single application.
򐂰 Data Cataloging as an IBM Storage Fusion service
– IBM Storage Fusion Data Cataloging service is available in IBM Storage Fusion. It is a
container native modern metadata management software that 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 service, see Data Cataloging.
򐂰 New Backup and Restore IBM Storage Fusion service
– The new Backup & Restore service in IBM Storage Fusion allows you to backup and
restore applications. It includes the following additional benefits in comparison with
Backup & Restore (Legacy):
• Policy frequency can be realized in units of minutes or hours with the new service
• Multiple policies can be assigned to single application.
• There is more visibility of job progress and error information with the new service.
• Greater resiliency and scale-out capabilities.
– For more information about the service, see Backup and Restore.

򐂰 Add IBM Storage Protect as a backup storage location through S3 option
– IBM Storage Fusion allows you to backup and recover Red Hat OpenShift data to an
IBM Storage Protect server with the S3 option. For more information about backing up
to IBM Storage Protect, see Using IBM Storage Protect as a backup storage location.
򐂰 New supported component versions in IBM Storage Fusion
– Red Hat OpenShift Container Platform 4.12 for IBM Storage Fusion
– IBM Storage Fusion Data Foundation 4.12
– Data Cataloging 2.1.0
– Backup & Restore 2.5.1
– For supported IBM Cloud Paks versions, see IBM Cloud Paks support for IBM Storage
Fusion
4.14.3 New features and changes in 2.4

Listed here are some of the new features in IBM Spectrum Fusion 2.4:
򐂰 Support for the IBM Spectrum Fusion HCI introduces 64-core servers (9155-C05) as an
alternative to the existing 32-core servers.
򐂰 Improvements to Metro Sync DR
򐂰 Ability to deploy IBM Spectrum Fusion on IBM Cloud, Microsoft Azure, or zLinux. For more
information about these deployments, see IBM Spectrum Fusion deployments.
򐂰 Scaling out Red Hat OpenShift Data Foundation and detecting an existing Red Hat
OpenShift Data Foundation configuration. For more information about Red Hat OpenShift
Data Foundation service and storage configurations, see Red Hat Advanced Cluster
Management for Kubernetes.
򐂰 Mirroring IBM Spectrum Fusion images to your enterprise registry. For more information
about mirroring your images to your enterprise registry, see Enterprise registry for IBM
Spectrum Fusion HCI installation.
Product versions in IBM Spectrum Fusion HCI and IBM Spectrum Fusion:
򐂰 Red Hat OpenShift Data Foundation 4.10
򐂰 IBM Spectrum Scale 5.1.6
򐂰 IBM Spectrum Scale Container Native 5.1.6 for IBM Spectrum Fusion HCI
򐂰 On-premises VMware supported version is vSphere 7.0
򐂰 IBM Spectrum Protect Plus 10.1.13
򐂰 IBM Spectrum Discover 2.0.6
For supported IBM Cloud Paks versions, see IBM Cloud Paks support for IBM Spectrum
Fusion.

Listed here are some of the new features in IBM Spectrum Fusion 2.3:
򐂰 IBM Spectrum Fusion HCI introduces 64-core servers (9155-C04) as an alternative to the
existing 32-core servers.
򐂰 Upgrade memory for the new 64-core servers.
– These servers come with base memory of 1024 GB of RAM upgradable to 2048 GB of
RAM.
򐂰 Mirror images to multiple registries
– IBM Spectrum Fusion supports installation from multiple enterprise registries. It gives
you the flexibility to mirror images to individual registries of your choice.
򐂰 IBM Spectrum Fusion HCI Metro Sync DR (Disaster Recovery) between two IBM
Spectrum Fusion HCI OpenShift Container Platform clusters.
– IBM Spectrum Fusion HCI provides a unique active-active resiliency across data
centers that are up to 150km distance.
򐂰 For more information about Metro Sync DR, see
– Metro Sync DR
򐂰 Upgrade of IBM Spectrum Fusion HCI node and switch firmware
– To know more about compute node and switch firmware upgrades, see here:
– Upgrading the compute nodes
– Upgrading the switch firmware
򐂰 Encryption option is available in the user interface for IBM Spectrum Fusion HCI. For more
details about the user interface page, see
– Setting encryption for storage
򐂰 Call Home support is included in IBM Spectrum Fusion and you can enable it from the IBM
Spectrum Fusion user interface. For steps to enable from the user interface, see
– Enabling Call Home
򐂰 Serviceability Improvements
򐂰 Supported product versions
– Red Hat OpenShift Container Platform 4.10
– IBM Spectrum Scale 5.1.5
– On-premises VMware supported version is vSphere 7.0
– IBM Spectrum Protect Plus 10.1.12
– IBM Spectrum Discover 2.0.5 is available in the Operator Hub catalog as a Container
software.
򐂰 For a more detailed listing of all the new features in IBM Spectrum Fusion 2.3, see the link
here:
– New Features and Changes in IBM Spectrum Fusion 2.3

Listed here are the new features in IBM Spectrum Fusion 2.2.1:
򐂰 This release includes defect fixes and security enhancements for IBM Spectrum Fusion
and IBM Spectrum Fusion HCI.
򐂰 Setup IBM Spectrum Fusion HCI in a restricted network. For more information about
mirroring your images to the enterprise registry, see Mirroring your images to the
enterprise registry.
Supported product versions:

򐂰 List of product versions included in 2.2.1:
– Red Hat OpenShift Container Platform 4.10 for IBM Spectrum Fusion. This support is
not applicable for IBM Spectrum Fusion HCI.
– IBM Spectrum Protect Plus 10.1.10.2
For more information see link here:
New Features and Changes in IBM Spectrum Fusion 2.2.1

Listed here are the new features in IBM Spectrum Fusion 2.2:
򐂰 Converged user interface for IBM Spectrum Fusion HCI and IBM Spectrum Fusion
򐂰 Support for IBM Cloud Paks
򐂰 Application backup using IBM Spectrum Fusion
IBM Spectrum Fusion HCI

򐂰 Enable purchase of network transceivers, cables, and spine switch
򐂰 Memory expansion options
򐂰 Support to upgrade IBM Spectrum Fusion HCI components
IBM Spectrum Fusion

򐂰 IBM Spectrum Fusion operator
򐂰 Storage options for IBM Spectrum Fusion
New Features and Changes in IBM Spectrum Fusion 2.2

Listed here are the new features in IBM Spectrum Fusion 2.1.2
򐂰 Enhanced Active file management (AFM)
򐂰 Container Network Interface (CNI) daemon network
򐂰 Support for Red Hat OpenShift Container Platform
򐂰 Support for IBM Spectrum Scale Erasure Code Edition (ECE)

򐂰 Support for IBM Spectrum Protect Plus
򐂰 Support for proxy server
New Features and Changes in IBM Spectrum Fusion 2.1.2.
4.14.8 IBM Storage Fusion Useful Links

򐂰 IBM Storage Fusion Documentation Main Page
https://www.ibm.com/docs/en/storage-fusion
򐂰 IBM Storage Fusion 2.5 Support Portal
https://www.ibm.com/support/pages/node/6966946
򐂰 IBM Spectrum Fusion 2.4 Support Portal
򐂰 IBM Spectrum Fusion Passport Advantage®
https://www.ibm.com/software/passportadvantage

Related publications
The publications listed in this section are considered particularly suitable for a more detailed
discussion of the topics covered in this paper.
IBM Redbooks
The following IBM Redbooks publications provide additional information about the topic in this
document. Note that some publications referenced in this list might be available in softcopy
only.
򐂰 IBM Storage Fusion HCI System Metro Sync Dr Use Case, REDP-5708
򐂰 IBM Storage Fusion Backup and Restore for Cloud Pak for Data, REDP-5706
򐂰 IBM Spectrum Scale Security, REDP-5426
򐂰 Multi-Factor Authentication Using IBM Security Verify for IBM Spectrum Fusion,
REDP-5662
You can search for, view, download or order these documents and other Redbooks,
Redpapers, Web Docs, draft and additional materials, at the following website:
ibm.com/redbooks
Online resources
These websites are also relevant as further information sources:
򐂰 IBM Storage Fusion documentation
https://www.ibm.com/docs/en/storage-fusion
򐂰 IBM Spectrum Fusion Passport Advantage
https://www.ibm.com/software/passportadvantage
򐂰 IBM Storage Modeller
https://www.ibm.com/tools/storage-modeller/help/index.html#page/WWR2/5528_intro
.IBM_Storage_Modeller.html
򐂰 Metro-DR
https://www.ibm.com/docs/en/storage-fusion/2.5?topic=prerequisites-metro-dr-dis
aster-recovery
򐂰 Red Hat OpenShift Container Platform security and compliance
https://docs.openshift.com/container-platform/4.11/security/index.html

Help from IBM
IBM Support and downloads
ibm.com/support
IBM Global Services

ibm.com/services

Back cover
REDP-5688-00
ISBN 0738460990
Printed in U.S.A.
®
ibm.com/redbooks

Redp 5688

Uploaded by

Copyright:

Available Formats

Redp 5688

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Redp 5688

Uploaded by

Copyright:

Available Formats

Front cover

IBM Storage Fusion Product Guide

IBM Storage Fusion Product Guide

First Edition (June 2023)

© Copyright International Business Machines Corporation 2023. All rights reserved.

Chapter 1. Why containers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1

Chapter 2. Why IBM Storage Fusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11

Chapter 3. Flavors of IBM Storage Fusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25

Chapter 4. IBM Storage Fusion Terminology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39

© Copyright IBM Corp. 2023. iii

iv IBM Storage Fusion Product Guide

INTERNATIONAL BUSINESS MACHINES CORPORATION PROVIDES THIS PUBLICATION “AS IS”

© Copyright IBM Corp. 2023. v

The following terms are trademarks of other companies:

vi IBM Storage Fusion Product Guide

Chapter 4, “IBM Storage Fusion Terminology” on page 39 describes the terminology

© Copyright IBM Corp. 2023. vii

Thanks to the following people for their contributions to this project:

viii IBM Storage Fusion Product Guide

Stay connected to IBM Redbooks

June 2023, First Edition (minor update)

© Copyright IBM Corp. 2023. xi

Chapter 1. Why containers

© Copyright IBM Corp. 2023. 1

Figure 1-1 Pets versus Cattle1

1 images: creative commons CC-0

2 IBM Storage Fusion Product Guide

1.2 Faster time to market in IT

A real-world case: Medical support platform in 3 weeks

The container advantage

Chapter 1. Why containers 3

IBM Cloud Paks

Swift deployment of IBM Cloud Paks on IBM Storage Fusion

https://app.ibmsalesconfigurator.com (requires an IBM account and a public IBMid)

4 IBM Storage Fusion Product Guide

The concept of “describing all in one place” is beneficial when it comes to

In a containerized environment, these two subjects are barely distinguishable.

Building Infrastructure with Developer Tools

The deployment process becomes increasingly hardware agnostic or “Cloud-inspired”,

1.4 Cyber Security and Cyber Resiliency

Apache Log4j lessons learned

Chapter 1. Why containers 5

Containerized environments simplify frequent recovery testing by design: There is no

Infrastructure-as-code simplifies the deployment in another location with a compatible control

Zero-trust for dynamic environments

6 IBM Storage Fusion Product Guide

1.6 Workload Scalability

Table 1-1 Classic versus containerized performance management approach

Scaling Storage for Applications

Chapter 1. Why containers 7

The two supported storage options are, at the time of writing:

1.7 Deconstruction silos with Data Bridging

IBM Storage Scale can:

8 IBM Storage Fusion Product Guide

Chapter 1. Why containers 9

Chapter 2. Why IBM Storage Fusion

© Copyright IBM Corp. 2023. 11

In the following sections we will explore these benefits in more detail.

2.2 Why Red Hat OpenShift on Bare Metal