Storwize and IBM I PDF

Front cover
IBM i and IBM Storwize Family

A Practical Guide to Usage Scenarios
VIOS explained from an IBM i

perspective
Using PowerHA together with IBM

Storwize Family
Configuring BRMS in a high

available solution
Sabine Jordan
Mario Kisslinger
Rodrigo Jungi Suzuki
ibm.com/redbooks
International Technical Support Organization
IBM i and IBM Storwize Family: A Practical

Guide to Usage Scenarios
April 2014
SG24-8197-00
Note: Before using this information and the product it supports, read the information in “Notices” on
page vii.
First Edition (April 2014)
This edition applies to IBM i 7.1 Technology Refresh (TR) 7.
© Copyright International Business Machines Corporation 2014. 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 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . vii
Trademarks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . viii
Preface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
Authors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
Now you can become a published author, too! . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
Comments welcome. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
Stay connected to IBM Redbooks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
Chapter 1. Introduction to IBM Storwize Family for IBM i . . . . . . . . . . . . . . . . . . . . . . . . 5

1.1 Decision support for internal versus external storage . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
1.2 Attachment options . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
1.2.1 Direct attached . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
1.2.2 Native attached . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
1.2.3 vSCSI attached . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
1.2.4 N-Port ID Virtualization (NPIV) attached . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
1.3 Basics of IBM Storwize family . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
1.4 IBM Storwize V7000 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
1.4.1 Control enclosure models . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
1.4.2 Expansion enclosure models . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15
Chapter 2. Configuring IBM Storwize V7000 with IBM i. . . . . . . . . . . . . . . . . . . . . . . . . 17

2.1 Initial setup . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18
2.2 Initial configuration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19
2.3 Preparing the IBM Storwize V7000 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25
2.3.1 Step 1: Creating the storage pool . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25
2.3.2 Step 2: Creating a generic volume . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29
2.3.3 Step 3: Creating a compressed volume . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31
2.3.4 Step 4: Creating a host . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34
2.3.5 Step 5: Mapping the LUN to the host . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36
2.4 IBM Storwize V7000 and hardware compression . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37
2.4.1 Using the IBM Comprestimator Utility . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38
2.4.2 Creating a volume mirrored copy on a compressed volume. . . . . . . . . . . . . . . . . 39
2.4.3 Comparing the performance of compressed and non-compressed volumes . . . . 43
2.5 Tips for implementing IBM Storwize V7000 in a IBM i environment . . . . . . . . . . . . . . . 50
2.5.1 Storwize V7000 configuration options. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 50
2.5.2 Host attachment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 51
2.5.3 Multipath . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 52
2.5.4 Zoning SAN switches . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 55
2.5.5 Data layout . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 57
2.5.6 Solid-State Drives . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 59
2.5.7 Booting from SAN . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 60
2.5.8 Sizing for performance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 60
2.5.9 Performance monitoring . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 61
2.5.10 Copy services considerations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 61
2.5.11 Translation from 520-byte blocks to 512-byte blocks . . . . . . . . . . . . . . . . . . . . . 62
Chapter 3. Setting up VIOS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 63

3.1 Before you start . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 64
© Copyright IBM Corp. 2014. All rights reserved. iii

3.2 VIOS setup by using HMC GUI . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 65
3.3 VIOS setup by using the CLI . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 69
3.4 Mapping storage through VIOS to IBM i . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 72
3.4.1 Setting up vSCSI connections . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 72
3.4.2 Mapping storage by using NPIV . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 76
3.5 Network setup . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 79
3.5.1 Optional step: LACP setup for adapter redundancy in each VIOS . . . . . . . . . . . . 80
3.5.2 Creating VLAN-capable trunk Ethernet adapters . . . . . . . . . . . . . . . . . . . . . . . . . 80
3.5.3 Shared Ethernet Adapter setup . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 83
3.5.4 Transporting VLANs to the external network . . . . . . . . . . . . . . . . . . . . . . . . . . . . 85
3.6 Customizing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 85
3.6.1 Changing ksh to bash style . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 86
3.6.2 Configuring monitoring by using SNMP . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 86
3.6.3 Creating an image repository for client D-IPL . . . . . . . . . . . . . . . . . . . . . . . . . . . . 87
3.7 Troubleshooting . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 92
3.7.1 Measuring performance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 92
3.7.2 Restarting the RMC connection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 95
3.7.3 Collecting troubleshooting data for support. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 96
Chapter 4. PowerHA and the IBM Storwize Family . . . . . . . . . . . . . . . . . . . . . . . . . . . . 97

4.1 Usage scenarios for IBM PowerHA SystemMirror for i . . . . . . . . . . . . . . . . . . . . . . . . . 98
4.1.1 Metro Mirror . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 99
4.1.2 Global Mirror . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 99
4.1.3 LUN level switching. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 100
4.1.4 FlashCopy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 102
4.2 Setting up IBM PowerHA SystemMirror for i . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 103
4.2.1 Requirements for PowerHA SystemMirror for i with SAN Volume Controller or
Storwize . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 103
4.2.2 SAN Volume Controller split cluster and LUN level switching. . . . . . . . . . . . . . . 105
4.2.3 LUN level switching in combination with Global Mirror . . . . . . . . . . . . . . . . . . . . 137
4.2.4 Metro Mirror setup. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 150
4.2.5 Adding LUNs to an existing IASP . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 150
Chapter 5. Live Partition Mobility . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 157

5.1 Usage scenarios for Live Partition Mobility . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 158
5.2 License considerations with Live Partition Mobility . . . . . . . . . . . . . . . . . . . . . . . . . . . 162
5.3 Setting up Live Partition Mobility . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 163
5.3.1 Requirements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 163
5.3.2 Preparation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 164
5.3.3 Live Partition Mobility migration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 168
5.3.4 Hints and tips for the validation process . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 170
Chapter 6. BRMS and FlashCopy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 173

6.1 Introduction to PowerHA, BRMS, and FlashCopy scenario . . . . . . . . . . . . . . . . . . . . 174
6.2 PowerHA and FlashCopy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 174
6.2.1 Preparation for FlashCopy on the Storwize side. . . . . . . . . . . . . . . . . . . . . . . . . 174
6.2.2 Changes in PowerHA to add the third node into the cluster . . . . . . . . . . . . . . . . 175
6.3 Backup Recovery, and Media Services (BRMS). . . . . . . . . . . . . . . . . . . . . . . . . . . . . 176
6.3.1 Setting up the BRMS network group. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 176
6.3.2 Configuring BRMS media classes and media policies . . . . . . . . . . . . . . . . . . . . 178
6.3.3 Configuring the BRMS control group . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 179
Appendix A. PowerHA Tools IBM i. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 183

PowerHA Tools for IBM i . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 184
iv IBM i and IBM Storwize Family: A Practical Guide to Usage Scenarios

IBM Lab Services Offerings for PowerHA for IBM i . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 185
Related publications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 187

IBM Redbooks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 187
Other publications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 187
Online resources . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 187
Help from IBM . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 188
Contents v
vi IBM i and IBM Storwize Family: A Practical Guide to Usage Scenarios
Notices
This information was developed for products and services offered in the U.S.A.
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, Armonk, NY 10504-1785 U.S.A.
The following paragraph does not apply to the United Kingdom or any other country where such
provisions are inconsistent with local law: 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 states 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 supply in any way it believes appropriate without incurring
any obligation to you.
Any performance data contained herein was determined in a controlled environment. Therefore, the results
obtained in other operating environments may vary significantly. Some measurements may have been made
on development-level systems and there is no guarantee that these measurements will be the same on
generally available systems. Furthermore, some measurements may have been estimated through
extrapolation. Actual results may vary. Users of this document should verify the applicable data for their
specific environment.
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.
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 the names and addresses used by an actual business
enterprise 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.
© Copyright IBM Corp. 2014. All rights reserved. vii

Trademarks
IBM, the IBM logo, and ibm.com are trademarks or registered trademarks of International Business Machines
Corporation in the United States, other countries, or both. These and other IBM trademarked terms are
marked on their first occurrence in this information with the appropriate symbol (® or ™), indicating US
registered or common law trademarks owned by IBM at the time this information was published. Such
trademarks may also be registered or common law trademarks in other countries. A current list of IBM
trademarks is available on the Web at http://www.ibm.com/legal/copytrade.shtml
The following terms are trademarks of the International Business Machines Corporation in the United States,
other countries, or both:
AIX® Informix® Redbooks®
DB2® Power Systems™ Redbooks (logo) ®
DS8000® POWER7® Storwize®
Easy Tier® PowerHA® System Storage®
FlashCopy® PowerVM® SystemMirror®
IBM® Real-time Compression™ Tivoli®
The following terms are trademarks of other companies:
ITIL is a registered trademark, and a registered community trademark of The Minister for the Cabinet Office,
and is registered in the U.S. Patent and Trademark Office.
Microsoft, Windows, and the Windows logo are trademarks of Microsoft Corporation in the United States,
other countries, or both.
UNIX is a registered trademark of The Open Group in the United States and other countries.
Other company, product, or service names may be trademarks or service marks of others.
viii IBM i and IBM Storwize Family: A Practical Guide to Usage Scenarios
Preface
The use of external storage and the benefits of virtualization became a topic of discussion in
the IBM® i area during the last several years. The question tends to be, what are the
advantages of the use of external storage that is attached to an IBM i environment as
opposed to the use of internal storage. The use of IBM PowerVM® virtualization technology
to virtualize Power server processors and memory also became common in IBM i
environments. However, virtualized access to external storage and network resources by
using a VIO server is still not widely used.
This IBM Redbooks® publication gives a broad overview of the IBM Storwize® family
products and their features and functions. It describes the setup that is required on the
storage side and describes and positions the different options for attaching IBM Storwize
family products to an IBM i environment. Basic setup and configuration of a VIO server
specifically for the needs of an IBM i environment is also described.
In addition, different configuration options for a combined setup of IBM PowerHA®

SystemMirror® for i and the Storwize family products are described and positioned against
each other. Detailed examples are provided for the setup process that is required for these
environments.
The information that is provided in this book is useful for clients, IBM Business Partners, and
IBM service professionals who need to understand how to install and configure their IBM i
environment with attachment to the Storwize family products.
Authors
This book was produced by a team of specialists from around the world working at the
International Technical Support Organization, Rochester Center.
Sabine Jordan is a Consulting IT Specialist in IBM Germany.

She has more than 20 years experience as a Technical
Specialist in the IBM i, specializing in high availability. Currently
she works on IBM PowerHA SystemMirror for i
implementations for SAP and non-SAP environments that use
geographic mirroring, IBM DS8000®, and SAN Volume
Controller remote copy services. She creates concepts for the
design and implements the entire project (cluster setup,
application changes), in addition to performing customer
education and testing. Sabine also presents and delivers
workshops (internally and externally) on IBM PowerHA
SystemMirror for i, high availability, and disaster recovery.
© Copyright IBM Corp. 2014. All rights reserved. 1

Mario Kisslinger is a Solution Architect for UBL
Informationssysteme GmbH in Germany. He has six years of
experience with designing infrastructure solutions for IBM i
customers. In complex project situations, he accompanies the
whole project lifecycle. As a Pre-Sales consultant, he designs
and configures the solution and supports the proposal. In the
implementation phase, he installs the solution. He also helps
with the transition into the operation phase. His areas of
expertise include backup and recovery, migration scenarios,
and serving IBM i out of the datacenter.
Rodrigo Jungi Suzuki is SAN Storage specialist at IBM Brazil

Global Delivery Center in Hortolandia. Currently, Rodrigo is a
Subject Matter Expert (SME) account focal point, and works
with projects, implementations, and support for international
clients. He has 20 years of IT Industry experience with the last
five years in the SAN Storage area. He has a background in
UNIX and Informix® database. He holds a bachelor degree in
Computer Science from Universidade Paulista in Sao Paulo,
Brazil, and is an IBM Certified IT Specialist. Rodrigo also is
certified in NetApp NCDA, IBM Storwize V7000 Technical
Solutions V2, and ITIL.
This edition of this IBM Redbooks project was led by Debbie Landon of the International
Technical Support Organization, Rochester Center.
Thanks to the following people for their contributions to this project:
David Bhaskaran
Jenny Dervin
Steven Finnes
Brian Nordland
Marilin Rodriguez
Kiswanto Thayib
Kristopher Whitney
Keith Zblewski
IBM Rochester, IBM i Development team
Ingo Dimmer
IBM Germany, Advanced Technical Skills (ATS) team
David Painter
IBM United Kingdom, STG Lab Services
Mike Schambureck
IBM US, IBM i Technology Center (iTC)
Alison Pate
IBM US, Advanced Technical Skills (ATS) team
Ann Lund
International Technical Support Organization
2 IBM i and IBM Storwize Family: A Practical Guide to Usage Scenarios

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 ITSO 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
this website:
http://www.ibm.com/redbooks/residencies.html
Comments welcome
Your comments are important to us!
We want our books to be as helpful as possible. Send us your comments about this book or
other IBM Redbooks publications in one of the following ways:
򐂰 Use the online Contact us review Redbooks form that is found at this website:
http://www.ibm.com/redbooks
򐂰 Send your comments in an email to:
[email protected]
򐂰 Mail your comments to:
IBM Corporation, International Technical Support Organization
Dept. HYTD Mail Station P099
2455 South Road
Poughkeepsie, NY 12601-5400
Stay connected to IBM Redbooks

򐂰 Find us on Facebook:
http://www.facebook.com/IBMRedbooks
򐂰 Follow us on Twitter:
http://twitter.com/ibmredbooks
򐂰 Look for 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 3
1
Chapter 1. Introduction to IBM Storwize

Family for IBM i
With the entrance of the IBM Storwize Family into the market, many options to attach external
storage to an IBM i host system became available. This chapter describes the pros and cons
of external versus internal storage. This chapter also describes the available options to
connect storage of the IBM Storwize Family to IBM Power Systems and explains the
differences between the available Storwize models.
The following topics are covered in this chapter:

򐂰 Decision support for internal versus external storage
򐂰 Attachment options
򐂰 Basics of IBM Storwize family
򐂰 IBM Storwize V7000

1.1 Decision support for internal versus external storage
IBM i clients focus on a high number of fast disk arms to handle the small I/O block sizes that
often are on IBM i. Disk performance is always mission critical.
It is reasonable to use internal storage for the following reasons:

򐂰 Easy to distribute specific physical disk arms between the production, test, and
development partitions.
򐂰 Everything is managed from one box.
򐂰 In case of errors, there is only one contact person in support.
򐂰 Easy to set up within one department.
򐂰 There are no unwanted side effects if other partitions produce a high workload.
However, internal disks can be annoying for the following reasons:

򐂰 Low flexibility to expand or reduce the number of disks if needed.
򐂰 Complete disks or expansion units must be provisioned to partitions.
򐂰 No load balancing between disks with high read/write usage and those with low read/write
usage.
򐂰 The IBM i administrator must worry about disk protection and the number of hot spare
disks.
External storage helps solve the annoying points of internal storage and can provide solutions
for most of the positive points. External storage features the following advantages:
򐂰 Storage area network (SAN) storage can be placed in different pools for production and
test workloads. It can also be implemented in different storage tier; for example, with
different reliability and performance demands.
򐂰 Availability and disk protection is managed by the storage administrator.
򐂰 It is much easier to move or replicate the storage into other areas to minimize the risk of
disaster.
򐂰 Storage can be segmented into more and smaller parts, with the flexibility to move storage
between partitions.
򐂰 Smaller footprint in the data center that leads to reduced power and cooling costs.
򐂰 Investment cycle for storage and server can be split apart, which allows longer use of the
external storage when the IBM Power System is replaced.
As a rule, the advantages to use external storage are growing, as more partitions must be
deployed. Every logical disk unit is distributed to many physical disks and automatically
protected against failure. It is possible to set up new workloads when they are needed,
without the demand for more disk adapters or storage boxes.
With the support for external storage in IBM PowerHA SystemMirror for IBM i or the built-in
replication functions of the IBM Storwize Family, a broad range of disaster recovery and high
availability options are available.
IBM PowerVM Live Partition Mobility also depends on external storage. With this product,
logical partitions can be moved to another IBM Power System, without the need for
downtimes. This can be used for planned outages like maintenance windows or migrations.

1.2 Attachment options
There are several different options to connect external storage to the IBM i operating system.
This section explains the differences between these attachment options and provide some
hints about when to use which option.
For more information about possible combinations of Host Bus Adapters, specific IBM Power
Systems, and firmware levels, see the IBM System Storage® Interoperation Center (SSIC) at
this website:
http://www-03.ibm.com/systems/support/storage/ssic/interoperability.wss
1.2.1 Direct attached

Figure 1-1 shows that direct attached means a connection from a member of the IBM
Storwize Family to a logical partition that is running IBM i without a switch. It is important to
connect each Host Bus Adapter to each Node of the storage system for redundancy.
Figure 1-1 Direct attachment of IBM Storwize V7000 to IBM Power System
Important: Direct attached is restricted to the 4 GB adapters, 5774 or the low profile
adapter 5276, and requires IBM i 7.1 TR6 with the latest Program Temporary Fixes (PTFs).
Direct attached storage is a good choice for systems with up to two logical partitions. The
number of partitions is limited by the Fibre Channel Ports in the storage system. It is possible
to use fewer Host Bus Adapters at the client partition, but this limits the number of active
paths for the disks to one, which is not considered a best practice and can be error prone.
Chapter 1. Introduction to IBM Storwize Family for IBM i 7

1.2.2 Native attached
As shown in Figure 1-2, in a native attached setup, the storage is connected to IBM i by using
a SAN switch. IBM i 7.1 TR6 with the latest PTFs is the minimum requirement for this
attachment option.
Figure 1-2 Two native attached IBM Storwize V7000 to two IBM Power Systems
The native attach option allows for a greater number of IBM i partitions (compared to the
direct attach option) because all Host Bus Adapters are connected to the switch. The amount
of the maximum partitions is limited by the number of SAN switch ports and slots in the IBM
Power System.
1.2.3 vSCSI attached

For a storage attachment that uses vSCSI, an IBM Virtual I/O Server (VIOS) must be installed
(see Figure 1-3 on page 9). The logical disks from shared SAN storage are assigned to the
VIOS partitions. The VIOS provides them to the IBM i client by using virtual SCSI
connections. For more information about the necessary steps for this configuration, see 3.4.1,
“Setting up vSCSI connections” on page 72.
The use of VIOS as virtualization layer provides a highly flexible and customized solution
where it is possible to use fewer physical adapters as compared to direct attached and native
attached configurations.
Disks can be attached directly to the VIOS partitions (without the use of a switch) or they can
be provided through the SAN fabric. In either case, the disks are mapped directly to the VIOS
where they are shown as hdiskX devices. Afterward, a mapping from VIOS to the IBM i client
must be configured.
Note: It is also possible to provide internal disks through VIOS to the IBM i partitions by
using vSCSI and mirroring these disks through the IBM i operating system for resilience.

Figure 1-3 Dual VIOS setup for vSCSI disks
1.2.4 N-Port ID Virtualization (NPIV) attached

Figure 1-4 on page 10 shows the difference between vSCSI and N-Port ID Virtualization
(NPIV) attached. Instead of mapping the disks to the VIOS, the disks are transported
transparently through VIOS and mapped to the IBM i client. For more information about the
setup process for this configuration, see 3.4.2, “Mapping storage by using NPIV” on page 76.
Virtual Fibre Channel client adapters are assigned to the IBM i partitions. The IBM Power
server hypervisor automatically generates a pair of virtual World Wide Port Numbers
(WWPNs) for this adapter that is used for SAN zoning and the host attachment within the
SAN storage.
Note: An NPIV capable Host Bus Adapter and an NPIV enabled SAN switch are required.
The transparent connection of NPIV allows for better performance than virtual SCSI,
simplified SAN setup, and multipath connections to the storage by adding them to different
zones, as described in 2.5.3, “Multipath” on page 52.
The NPIV setup is the most flexible configuration for a large number of partitions on a single
IBM Power System.

Figure 1-4 Dual VIOS setup for NPIV disks
1.3 Basics of IBM Storwize family

The SAN Volume Controller is a storage virtualization system that facilities storage
management, providing efficiency. It helps to improve storage space usage and SAN
administrator productivity.
The SAN Volume Controller can be a single point of control to provision LUNs to your entire
SAN environment. It can virtualize a variety of storage systems. For more information, see the
following IBM System Storage SAN Volume Controller website:
http://www-03.ibm.com/systems/storage/software/virtualization/svc/
You can use IBM SAN Volume Controller with two or more nodes, which are arranged in a
cluster. A cluster is a pair of nodes and can grow up to four SAN Volume Controller node
pairs.

Figure 1-5 shows the front view of an IBM SAN Volume Controller.
Figure 1-5 SAN Volume Controller front view
The IBM SAN Volume Controller main function is to virtualize the disk drives that are provided
from different storage systems that are connected at the back end. The main function of the
IBM Storwize V3700 and IBM Storwize V7000 is to provide disks from their expansion
enclosures. Although these are the most usual Storwize family usage situations, the IBM
Storwize V3700 and Storwize V7000 can also virtualize external storages.
One of the main differences between the IBM Storwize V3700 and the IBM Storwize V7000 is
capacity. The Storwize V3700 supports up to four expansion units and can have up to 240 TB
of capacity. The Storwize V7000 can support up to nine expansion units, reaching up to
480 TB of capacity.
Figure 1-6 shows the front view of Storwize V3700.
Figure 1-6 Storwize V3700 front view
The IBM Storwize V3700 is an entry-level storage system that is targeted to meet the
requirements of the small business market and the IBM Storwize V7000 is targeted toward
more complex environments. There are differences with the cache per controller, maximum
number of expansions, quantity of host interface, and so on.
If you need Real-time Compression™ function, that is supported on the IBM Storwize V7000.
For more information, see 2.4, “IBM Storwize V7000 and hardware compression” on page 37.

Figure 1-7 shows the front view of the IBM Storwize V7000. For more information about
configuring an IBM Storwize V7000 in an IBM i environment, see Chapter 2, “Configuring IBM
Storwize V7000 with IBM i” on page 17.
Figure 1-7 Storwize V7000 front view
There also is an IBM Storwize V3500 storage system. It consists of a one control enclosure.
The control enclosure contains disk drives and two nodes, which are attached to the SAN
fabric and reaches up to 36 TB of storage capacity.
Important: The IBM Storwize V3500 is available only in the following countries and
regions:
򐂰 People's Republic of China
򐂰 Hong Kong S.A.R. of the PRC
򐂰 Macao S.A.R. of the PRC
򐂰 Taiwan

The user interface for the IBM SAN Volume Controller is the same for all of the Storwize
family, as shown in Figure 1-8.
Figure 1-8 IBM SAN Volume Controller Overview panel
For more information about the features, benefits, and specifications of the IBM Storwize
family, see the following websites:
򐂰 IBM System Storage SAN Volume Controller:
򐂰 IBM Storwize V7000 and Storwize V7000 Unified Disk Systems:
http://www-03.ibm.com/systems/storage/disk/storwize_v7000/index.html
򐂰 IBM Storwize V3700:
http://www-03.ibm.com/systems/hk/storage/disk/storwize_v3500/index.html
Table 1-1 list a series of videos that are available for the IBM Storwize family.
Table 1-1 IBM Storwize videos

Video Description URL
Storwize Family http://www.youtube.com/watch?v=M0hIfLKUk1U
Storwize V3700 http://www.youtube.com/watch?v=IdxnH9YWF9s
Storwize V7000 http://www.youtube.com/watch?v=S2p4IQxIiec&feature=c4-

overview-vl&list=PL48FE37276F6014E8

Video Description URL
IBM Real-time Compression and http://www.youtube.com/watch?v=rgKj75kn2J0

Storwize V7000 and SAN Volume
Controller demonstration
1.4 IBM Storwize V7000

In this section, we provide more details about the IBM Storwize V7000.
An IBM Storwize V7000 system consists of machine type 2076 rack-mounted enclosures.
The control enclosure contains the main processing units that control the entire system and
can be connected to expansion enclosures. The expansion enclosures can hold up to 12
3.5-inch drives or up to 24 2.5-inch drives.
1.4.1 Control enclosure models

For the 2076-112 and 2076-124 models, there are two power supply slots numbered as 1 and
two canister slots numbered as 2 in Figure 1-9. The canister slots have Fibre Channel ports,
USB ports, Ethernet ports, and Serial-attached SCSI ports.
Machine type and model 2076-112 can hold up to 12 3.5-inch drives and the 2076-124 can
hold up to 24 2.5-inch drives.
Figure 1-9 Storwize V7000 control enclosure rear view, models 112 and 124
Figure 1-10 shows the rear view of the control enclosure for the machine type and models
2076-312 and 324, which includes 10 Gbps Ethernet ports capabilities.
Machine type and model 2076-312 can hold up to 12 3.5-inch drives and the 2076-324 can
hold up to 24 2.5-inch drives.
Figure 1-10 Storwize V7000 control enclosure rear view, models 312 and 324

1.4.2 Expansion enclosure models
Expansion enclosure model 2076-212 can hold up to 12 3.5-inch drives, as shown
Figure 1-11.
Figure 1-11 Storwize V7000 expansion enclosure front view, models 2076-112, 212, and 312
Expansion enclosure model 2076-224 can hold up to 24 2.5-inch drives, as shown

Figure 1-12.
Figure 1-12 Storwize V7000 expansion enclosure front view, models 2076-124, 224, and 324
The expansion enclosure has two power supply slots numbered as 1 and two canister slots
numbered as 2 as shown in Figure 1-13.
Figure 1-13 Storwize V7000 expansion enclosure rear view
Each canister has two SAS ports that must be connected if you are adding an expansion
enclosure.
Note: Control enclosures have Ethernet ports, Fibre Channel ports, and USB ports.
Expansion enclosures do not have these ports.

2
Chapter 2. Configuring IBM Storwize V7000

with IBM i
This chapter describes configuring an IBM Storwize V7000 with an IBM i system.

򐂰 Initial setup
򐂰 Initial configuration
򐂰 Preparing the IBM Storwize V7000
򐂰 IBM Storwize V7000 and hardware compression
򐂰 Tips for implementing IBM Storwize V7000 in a IBM i environment

2.1 Initial setup
This section describes how to perform the initial configuration of an IBM Storwize V7000.
The Storwize V7000 uses a USB key to start the initial configuration. The USB key comes
with the Storwize V7000 and contains a program that is called InitTool.exe to start the
initialization.
Tip: If you do not have the USB key that is shipped with the Storwize V7000, you can use
another USB key. You can also download the InitTool.exe file from this website:
http://www-933.ibm.com/support/fixcentral/swg/selectFixes?parent=Mid-range+disk
+systems&product=ibm/Storage_Disk/IBM+Storwize+V7000+(2076)&release=All&platfor
m=All&function=all
Complete the following steps to start the initial configuration of an IBM Storwize V7000:
1. Insert the USB key into a personal computer that is running the Microsoft Windows
operating system. The initialization program is started automatically if the system is
configured to autorun for USB keys or open the USB key from Windows system and click
InitTool.exe.
2. In the IBM Storwize V7000 Initialization Tool window (see Figure 2-1), select Initialize a
new system using the USB key and click Next.
Figure 2-1 IBM Storwize V7000 initialization tool
3. The initialization panel guides you through the following steps:

a. Enter the IP address, subnet mask, and network gateway address.
b. After the initialization process finishes, eject the USB key and insert it into any USB
port on the control enclosure of your Storwize V7000 system. After the USB key
indicator lights stops flashing, remove the USB key from the system.

c. Reinsert the USB key in your personal computer to continue setting up your system.
On the Microsoft Windows operating system, you see the results of the operation.
4. From a personal computer that has an Ethernet connection, open a supported web
browser and point to the system management IP address to start the graphical user
interface.
5. Log in as superuser and use passw0rd for the password. It is considered a best practice to
change the default password for security reasons.
Tip: Normally, the Storwize V7000 contains sufficient power for the system to start. If the
batteries do not have sufficient charge, the system cannot start. You must wait until the
system became available.
2.2 Initial configuration

Complete the following steps to complete the initial Storwize V7000 configuration:
1. From the Microsoft Windows operating system, open a web browser and point to the
system management IP address. You see the IBM Storwize V7000 GUI, as shown in
Figure 2-2.
Log in as superuser and use passw0rd for the password. Click Continue.
Figure 2-2 IBM Storwize V7000 initial configuration
Chapter 2. Configuring IBM Storwize V7000 with IBM i 19

2. Read and agree to the license agreement, as shown in Figure 2-3. Click Next.
Figure 2-3 IBM Storwize V7000 license agreement
3. Enter the system name, time zone, date, and time, as shown in Figure 2-4. Click Next.
Figure 2-4 IBM Storwize V7000 system name, time zone, and date setup

4. You can add a license for external storage, remote copy, and real-time compression, as
shown in Figure 2-5. Click Next.
Figure 2-5 IBM Storwize V7000 system license
5. Configure call home and email alert.

You can configure your system to send an email report to IBM support if a hardware issue
is detected. This function is called Call Home. To configure the Call Home feature,
complete the following steps:
a. Click Configure Email Event Notifications, as shown in Figure 2-6.
Figure 2-6 Email event notifications setup

b. Configure the email account information as shown in Figure 2-7. Click Next.
Figure 2-7 Email account information
c. Configure the email servers, as shown in Figure 2-8. Click Next.
Figure 2-8 Configuring the email servers
d. Enter the IBM support email address to receive the notifications, as shown in
Figure 2-9. Click Next.
Figure 2-9 Support notifications email address

e. Verify the Summary information as shown in Figure 2-10 to ensure the information that
is provided is correct. Click Finish.
Figure 2-10 Support notification summary
f. After the support notification is configured, you can edit the configuration by clicking
Edit or you can continue with the setup by clicking Next, as shown in Figure 2-11.
Figure 2-11 Support notification information

6. The next window displays the hardware configuration, as shown in Figure 2-12. Click
Next.
Figure 2-12 IBM Storwize V7000 hardware
7. An Add Enclosure pop-up window opens, as shown in Figure 2-13. Wait until the task is
finished and then click Close.
Figure 2-13 Adding an enclosure

8. The configuration wizard detects the internal disk drives, as shown in Figure 2-14. The
following choices are available:
– Accept the configuration that is proposed by selecting the check box to automatically
configure the internal storage and click Finish.
Note: All available disks are configured into RAID sets and hot spare drives that are
based on the number of drives that are found. However, this might not be what you
intended to configure. Check the intended configuration before you click Finish.
– Leave the check box cleared, click Finish, and continue with the storage configuration.
Figure 2-14 Internal disk drives configuration proposed
9. After you click Finish, wait until the setup Wizard is complete, as shown in Figure 2-15.
Figure 2-15 Wizard setup complete
2.3 Preparing the IBM Storwize V7000

This section describes the steps to set up the IBM Storwize V7000, which consists of the
following tasks:
򐂰 2.3.1, “Step 1: Creating the storage pool” on page 25
򐂰 2.3.2, “Step 2: Creating a generic volume” on page 29
򐂰 2.3.3, “Step 3: Creating a compressed volume” on page 31
򐂰 2.3.4, “Step 4: Creating a host” on page 34
򐂰 2.3.5, “Step 5: Mapping the LUN to the host” on page 36
2.3.1 Step 1: Creating the storage pool

Complete the following steps to create the storage pool:
1. Open the GUI by entering the management IP address into a web browser’s URL field and
log in to the Storwize V7000, as shown Figure 2-16 on page 26.

Figure 2-16 IBM Storwize V7000 login panel
Figure 2-17 shows the Storwize V7000 overview window.
Figure 2-17 Storwize V7000 Overview panel

2. Go to the storage pool panel to create a storage pool. Click the Pools image button, as
shown Figure 2-18.
Figure 2-18 Selecting the IBM Storwize V7000 Pools panel
3. To create a storage pool, click the New Pool button. In the Create Pool window, enter the
pool name and click Next, as shown Figure 2-19.
Figure 2-19 Creating a pool

4. Click Detect MDisks to show you the MDisks that are to be added to the storage pool, as
shown in Figure 2-20.
Figure 2-20 Detecting MDisks
5. When the Discover Devices task completes, click Close, as shown in Figure 2-21.
Figure 2-21 Discovering devices
6. Select the MDisks and click Finish to create the storage pool.
The storage pool is created. The next steps describe how to create the volumes to be used by
the systems.

2.3.2 Step 2: Creating a generic volume
Complete the following steps to create a generic volume:
1. Click the Volumes image and then select Volumes, as shown in Figure 2-22.
Figure 2-22 Volumes options
2. In the Volumes window, click New Volume. The New Volume window is displayed, as
Figure 2-23 Creating a volume

3. Click Generic and select the pool name. Enter the volume name and size, then click
Create, as shown in Figure 2-24.
Figure 2-24 Creating a generic volume
4. Figure 2-25 shows the output of the volume creation task. Click Close.
Figure 2-25 Volume creation task output
5. In the Volumes window, you can see the new volume, as shown in Figure 2-26.
Figure 2-26 Volume information

6. Right-click the volume to see the volume details, as shown in Figure 2-27.
Figure 2-27 Displaying volume details
2.3.3 Step 3: Creating a compressed volume

This step is optional. IBM Storwize V7000 is designed to improve storage efficiency by
compressing data through supported real-time compression for block storage. With real-time
compression, you save a considerable amount of space. IBM Real-time Compression
operates as data is written to disk, which avoids the need to store uncompressed data when
waiting for compression. On the Storwize V7000, the IBM Real-time Compression feature is
licensed by enclosures; on the IBM SAN Volume Controller, it is licensed by storage space (in
terabytes).
For more information about hardware compression, see 2.4, “IBM Storwize V7000 and
hardware compression” on page 37.
Complete the following steps if you want to create a compressed volume:

1. Click the Volumes image and select the Volumes option, as shown Figure 2-28.
Figure 2-28 IBM Storwize V7000 volumes options
2. At the top of the Volumes window, click New Volume.

3. In the New Volume window, click the Compressed icon to create a Compressed volume,
as shown in Figure 2-29.
Figure 2-29 Creating a volume

4. As shown in Figure 2-30, select the pool and enter the volume name and size. Click
Create.
Figure 2-30 Creating a compressed volume
5. After the volume is created, click Close, as shown in Figure 2-31.
Figure 2-31 IBM Storwize V7000 Create Volumes task panel

6. In the Volume window, you see the new compressed volume, as shown in Figure 2-32.
Figure 2-32 IBM Storwize V7000 compressed volume
7. Right-click the new volume and select Properties to see the volume details, as shown in
Figure 2-33.
Figure 2-33 IBM Storwize V7000 compressed volume details
2.3.4 Step 4: Creating a host

Complete the following steps to create a host:
1. Click the Hosts image and select Hosts, as shown in Figure 2-34.
Figure 2-34 Hosts options
2. In the Hosts window, click New Host.

3. In the Create Host window, select Fibre Channel Host or iSCSI Host, as shown
Figure 2-35.
Figure 2-35 Create Host options
4. In this example, Fibre Channel Host was selected. Enter the host name and add the
WWPN. In the Advanced Settings section, select the settings that are related to your
Storwize V7000 I/O Group configuration and then select the Host Type. Click Create Host,
Figure 2-36 Creating a Fibre Channel host

5. After the host is created, click Close, as shown in Figure 2-37.
Figure 2-37 Create host task output
2.3.5 Step 5: Mapping the LUN to the host

Complete the following steps to map the LUN to the host:
1. Click the Volumes image and select Volumes, as shown in Figure 2-38.
Figure 2-38 Volumes options
2. In the Volumes window, right-click the volume and select Map to Host, as shown in
Figure 2-39.
Figure 2-39 Volume mapping

3. In the Modify Host Mappings window, select the host, select the volumes to be mapped,
and click the right arrow to move the volume from the Unmapped Volumes column to the
Volumes Mapped to the Host column. Click Map Volumes, as shown in Figure 2-40.
Figure 2-40 Mapping volumes
The volume is now mapped to the server. Ask the server administrator to verify that the new
disks are now recognized.
2.4 IBM Storwize V7000 and hardware compression

IBM Storwize V7000 is designed to improve storage efficiency by compressing data through
supported real-time compression for block storage. With real-time compression, you save a
considerable amount of space. IBM Real-time Compression is used with primary data and in
production environments with database and email systems. IBM Real-time Compression
operates as data is written to disk, which avoids the need to store uncompressed data when
you are waiting for compression.
On the Storwize V7000, the IBM Real-time Compression feature is licensed by enclosures; on
the IBM SAN Volume Controller, it is licensed by storage space (in terabytes).
Important: A system with compressed volumes dedicates processor and memory

resources for running the Real-time Compression function. Clients must evaluate the
impact to existing workloads on systems that are heavily used before the Real-time
Compression function is enabled.

IBM Storwize V7000 and SAN Volume Controller customers can evaluate the Real-time
Compression capability at no cost. Clients have 45 days to evaluate the Real-time
Compression function. At the end of this trial period, you must purchase the required licenses
for Real-time Compression or disable the function. For more information about this program,
see the IBM Real-time Compression Evaluation User Guide, which is available at this
website:
http://www-01.ibm.com/support/docview.wss?uid=ssg1S7003988&myns=s028&mynp=OCST3FR7
&mync=E
On average, an IBM i 7.1 system can achieve a data compression ratio of approximately 1:3,
which is about a 66% of compression savings. For an example, see 2.4.2, “Creating a volume
mirrored copy on a compressed volume” on page 39.
2.4.1 Using the IBM Comprestimator Utility

The IBM Comprestimator Utility is a host-based utility that can be used to provide an estimate
for the achievable capacity savings by using Storwize V7000 compression and
thin-provisioning. This utility is delivered as a binary executable file and is run from a host with
access to the Storwize V7000 volumes to be analyzed. It performs read operations only.
The Comprestimator tool can analyze and provide estimated compression rate for data. By
using this tool, clients can determine the data that can be a good candidate for Real-time
Compression. You can download this tool from this website:
http://www-304.ibm.com/support/customercare/sas/f/comprestimator/home.html
Important: The IBM Comprestimator Utility does not run on IBM i; therefore, it can be used
only with VIOS attached storage and not direct attached storage.
Example 2-1 shows the use of this utility on a PowerVMVirtual I/O Server to provide an
estimate for the compression savings for an IBM i load source volume that is mapped to the
Virtual I/O Server.
Example 2-1 Comprestimator output

# ./comprestimator -d /dev/hdisk1
Version: 1.2.01 (Build u0004)
Start time: 30/07/2012 03:50:58
Device name: /dev/hdisk1
Device size: 20.0 GB
Number of processes: 10
Exhaustive: no
Sample# | Device | Size(GB) | Compressed | Total | Total | Thin Provisioning | Compression | Compression
| Name | | Size(GB) | Savings(GB)| Savings(%) | Savings(%) | Savings(%) | Accuracy Range(%)
--------+------------+----------+------------+------------+------------+-------------------+-------------+------------------
3526 |/dev/hdisk1 | 20.0 | 3.9 | 16.1 | 80.6% | 22.9% | 74.9% | 4.7%
In this example of an IBM i load source of a 20 GB volume on the Storwize V7000, the utility
estimates a thin-provisioning savings of 22.9% and a compression savings of 74.9%. This
results in a compressed size of 3.9 GB, which is a total capacity savings of 80.6%. This result
is the required physical storage capacity that is predicted to be reduced by 80.6% as a result
of the combined savings from thin provisioning and compression.
Note: A Storwize V7000 compressed volume is always a thinly provisioned volume from an
architectural perspective when it was created with a real size of 100%. Therefore, the
reported compression savings are in addition to the savings that are provided by thin
provisioning, which stores areas of zeros with minimal capacity.

Thin provisioning
IBM i can use thin provisioning because it is not apparent to the server. However, you must
provide adequate hard disk drives (HDDs) to sustain the required performance regardless of
whether the capacity is required. Also, IBM i 7.1 does not pre-format LUNs so that initial
allocations can be thin-provisioned; therefore, there is no space reclamation and the
effectiveness of the thin provisioning can decline over time.
You also must ensure that you have sufficient disks that are configured to maintain
performance of the IBM i workload.
Thin provisioning might be more applicable to test or development environments.
2.4.2 Creating a volume mirrored copy on a compressed volume

IBM Storwize V7000 volume mirroring is required to migrate from a standard volume to a
compressed volume. The following steps show how to create a volume (VDisk) copy from an
existing non-compressed IBM i volume to a compressed volume for evaluating the new
Storwize V7000 real-time compression feature; for example, in terms of compression ratio
and performance:
1. From the V7000 GUI, right-click an existing non-compressed IBM i volume and select
Volume Copy Actions  Add Mirrored Copy to create a VDisk mirrored copy on a
compressed volume, as shown in Figure 2-41.
Figure 2-41 IBM Storwize V7000 adding a mirrored copy

2. In the Add Volume Copy window, select Compressed as the volume type and a pool for
the new VDisk copy. Click Add Copy to create a VDisk copy on a compressed volume, as
Figure 2-42 IBM Storwize V7000 adding a volume copy
As an alternative to the GUI procedure, the addvdiskcopy CLI command can be used for
creating a compressed mirrored VDisk from an existing VDisk. The following CLI command
shows an example of creating a compressed mirrored VDisk from an existing VDisk ID 10 in
mdiskgroup 1, which initially allocates 2% real capacity and uses the thin-provisioning auto
expand option:
svctask addvdiskcopy -mdiskgrp 1 -rsize 2% -autoexpand -compressed 10
The VDisk copy creation synchronization progress can be monitored from the Running Tasks
information in the Storwize V7000 GUI or with the CLI command that is shown in
Example 2-2, which also shows the estimated completion time.
Example 2-2 Compression progress

IBM_2076:V7000-ctr-03:admin>svcinfo lsvdisksyncprogress 10
vdisk_id vdisk_name copy_id progress estimated_completion_time
10 i7PFE2_SYSBAS_0 0 100
10 i7PFE2_SYSBAS_0 1 51 120625171149

In this example, VDisk 10 is a 20 GB LUN that is used as an IBM i 7.1 load source. From its
18,204 MB net capacity that is available to IBM i from the virtual SCSI disk unit, approximately
59.3% is used, as shown in Figure 2-43.
Work with Disk Status I7PFE2

10/16/13 12:02:19
Elapsed time: 00:00:00
Size % I/O Request Read Write Read Write %

Unit Type (M) Used Rqs Size (K) Rqs Rqs (K) (K) Busy
1 6B22 19088 59.3 .0 .0 .0 .0 .0 .0 0
2 6B22 19088 51.2 .0 .0 .0 .0 .0 .0 0
3 6B22 19088 51.2 .0 .0 .0 .0 .0 .0 0
4 6B22 19088 51.2 .0 .0 .0 .0 .0 .0 0
Command
===>
F3=Exit F5=Refresh F12=Cancel F24=More keys
Figure 2-43 IBM i Work with Disk Status display
From the audit log entry that is shown in Example 2-3, you can see that the mirrored copy
creation was started at 14:21, with an estimated completion time of 17:11. This process takes
about 2:50 hours to finish the 20 GB LUN that uses the default volume synchronization rate of
50%, which provides a synchronization data rate of approximately 2 MB.
Example 2-3 Audit log entry

IBM_2076:V7000-ctr-03:admin>svcinfo catauditlog -first 1
audit_seq_no timestamp cluster_user ssh_ip_address result res_obj_id action_cmd
819 120625142114 superuser 9.155.44.66 0 svctask addvdiskcopy
-mdiskgrp 1 -rsize 2% -warning 80 -autoexpand -compressed 10

You can review the VDisk volume properties by right-clicking the volume and selecting
Properties from the context menu. In the Volume Details window, select Show Details. A
compression savings for the IBM i load source disk unit of 65.5% is reported, as shown in
Figure 2-44. Therefore, the total savings (including thin-provisioning) are 100% • 3.64 GB /
20 GB = 81.8%.
Figure 2-44 IBM Storwize V7000 volume details
Change the synchronization rate to 100% by clicking Edit or by using the svctask chvdisk
–syncrate 100 <vdisk_id> command for the remaining three 20 GB LUNs from ASP1, which
are on a V7000 RAID 10 array with 8x 10k RPM SAS drives that are each used by IBM i by
51.2%. This results in the following VDisk mirror creation times:
򐂰 10 minutes for a single compressed volume with up to 16% V7000 RTC CPU usage
򐂰 10 minutes for two compressed volumes with up to 32% V7000 RTC CPU usage
On average, approximately 66% of compression savings was achieved, which is a data

compression ratio of approximately 1:3 for the IBM i 7.1 system as shown by the volume
compression savings information that is highlighted in Figure 2-45 on page 43.

Figure 2-45 IBM Storwize V7000 Volumes by host panel
When the first compressed volume is created, some of the Storwize V7000 and SAN Volume
Controller CPU cores and memory within the I/O group are reserved for use by the real-time
compression engine. On a four-core system with Storwize V7000 or SAN Volume Controller
CF8 nodes, three cores are reserved for the compression engine. On a six-core SAN Volume
Controller CG8 nodes system, four cores are reserved.
These resources are returned for general system use again when the last compressed
volume is deleted. It is considered a best practice to enable compression on four- or six-node
systems only if the current system CPU usage is less than 25% for a four-node system and
less than 50% for a six-node system. This can have significant performance implications to all
hosts that are served by the storage unit.
2.4.3 Comparing the performance of compressed and non-compressed

volumes
For comparing the performance of compressed and non-compressed volumes, an IBM i
database test workload was used, which performs read and write I/Os on a 32 GB large
DB2® database.
In the first scenario of using Storwize V7000 VDisk mirroring, both volume copies (that is, the
compressed and non-compressed copies), were on the same RAID 10 rank that uses 8x 300
GB 10k RPM SAS drives. For the second scenario with a split VDisk mirroring, each copy was
on its own dedicated RAID rank configuration.
Note: The small Storwize V7000 disk configuration that is used in this example was
saturated by the IBM i workload. Therefore, it is not representative of a properly sized IBM
i SAN storage configuration. However, this example still shows a compression performance
evaluation.

Scenario 1: Volume mirroring
This scenario shows the achieved disk I/O performance results for the IBM i database
workload with the mirrored VDisk configuration that was created as described earlier in this
section. With the non-compressed volume flagged as the primary copy, approximately 899
read I/Os per second and 1165 write I/Os per second were achieved with a corresponding
service time of 9 ms and 11 ms, as shown in Figure 2-46.
Figure 2-46 IBM Storwize V7000 performance for non-compressed primary mirrored VDisks
After the primary copy was switched, which designates the preferred copy of the mirrored
VDisk to read from, from the non-compressed volume to the compressed volume by using the
svctask chvdisk •primary <copy_ID> <vdisk_ID> command, you can see a significant
increase in read I/O performance with 4681 read I/Os per second and 1062 write I/Os per
second with a service time of 1 ms and 12 ms, as shown in Figure 2-47 on page 45.
Note: Depending on the host workload characteristics, a significant read I/O performance
improvement when compressed volumes are used can be achieved. Because of the
compression for cache read misses and write de-stages from cache to disk, less data must
be transferred.
This result can be verified by comparing the disk backend read I/O rate for the MDisks, which
dropped from 966 read I/Os per second for the non-compressed primary (see Figure 2-46) to
only 181 read IOs per second for the compressed primary, as shown in Figure 2-47 on
page 45.

Figure 2-47 IBM Storwize V7000 performance for compressed primary mirrored VDisks
Scenario 2: Split volume mirroring

As described in “Scenario 1: Volume mirroring” on page 44, which used VDisk mirroring to
create a compressed copy of a non-compressed volume, this can be seen as an interim
solution for testing the efficiency of the Storwize V7000 compression engine before deciding
to stay with independent (that is non-mirrored) compressed or non-compressed volumes.
In this section, a second scenario of comparing the Storwize V7000 disk I/O performance with
IBM i database workload for non-mirrored non-compressed versus non-mirrored compressed
volumes is reviewed. The previously mirrored volumes were split into two independent copies
by migrating one mirrored copy to another equally configured RAID rank (MDisk) in another
storage pool (mdiskgrp) by using the svctask migratevdisk command and by splitting the
mirror by using the svctask splitvdiskcopy command.
IBM i storage management uses I/O bundling to limit the amount of physical disk I/O. That is,
multiple logical I/Os can be coalesced into a single physical disk I/O with a larger transfer
size. The degree of IBM i I/O bundling also depends on the disk I/O service times. From this
perspective, the IBM i disk I/O transfer sizes can vary for different storage configurations such
that it is more accurate here to compare the achieved data rates (MBps) than the I/O
throughput (I/O per second).

Figure 2-48 shows the IBM i Work with Disk Status display for the case of the use of
non-compressed volumes.

10/16/13 14:08:39

1 6B22 19088 78.6 1564 26.2 1097 466.9 4.6 77.0 79
2 6B22 19088 74.0 1857 26.0 1302 555.3 4.5 76.4 80
3 6B22 19088 74.0 1851 26.0 1303 548.1 4.5 77.2 76
4 6B22 19088 74.0 1869 25.8 1302 566.7 4.5 74.7 78
Command
===>
Figure 2-48 Work with Disk Status display, non-compressed volumes
Figure 2-49 shows the Work with Disk Status display for compressed volumes, which shows
the IBM i I/O bundling effect with slightly smaller write transfers being used for the lower
performing non-compressed volumes.

10/16/13 14:08:39

1 6B22 19088 78.7 2310 23.3 1744 566.5 4.5 81.3 66
2 6B22 19088 74.1 2714 23.3 2047 667.6 4.5 81.1 69
3 6B22 19088 74.1 2710 23.4 2050 659.2 4.5 82.3 68
4 6B22 19088 74.1 2744 23.2 2048 695.7 4.5 78.3 68
Command
===>
Figure 2-49 Work with Disk Status display, compressed volumes

By using non-compressed volumes for the IBM i database workload, looking at the 7-minute
averages from Figure 2-48 on page 46, 5004 read I/Os per second (22 MBps) and 2137 write
I/Os per second (159 MBps) with services times of 2 ms and 5 ms were achieved, as shown
in Figure 2-50.
Figure 2-50 IBM Storwize V7000 performance for split mirror uncompressed volumes (IOPS)

This is compared to looking at the 5-minute averages from Figure 2-49 on page 46, in which
7889 read I/Os per second (35 MBps) and 2589 write I/Os per second (204 MBps) with 0 ms
and 4 ms was achieved with the compressed volumes, as shown in Figure 2-51.
Figure 2-51 IBM Storwize V7000 performance for split mirror compressed volumes (IOPS)
As described in “Scenario 1: Volume mirroring” on page 44, significant disk I/O performance
improvements are possible by using compression when the V7000/SAN Volume Controller
disk backend is already highly used. This is due to the compression in which less data must
be read and written from and to the disk backend, which causes a lower disk usage rate and,
therefore, better disk response times.
Looking at the V7000 disk backend data rates, you can see a significant reduction of the
backend workload from 78 MBps reads and 172 MB/ps for the case of non-compressed
volumes that are shown in Figure 2-52 on page 49. This is down to only 14 MBps reads and
16 ps writes for compressed volumes, as shown in Figure 2-53 on page 49.
However, as can also be seen from Figure 2-53 on page 49, with the system CPU utilization
reaching around 50% and the real-time compression CPU utilization peaking at 56%, a
proper sizing of the Storwize V7000 system with regards to CPU utilization as described in
2.4.2, “Creating a volume mirrored copy on a compressed volume” on page 39 becomes
more important when the compression feature is used.

Figure 2-52 IBM Storwize V7000 performance for split mirror non-compressed volumes (MBps)
Figure 2-53 IBM Storwize V7000 performance for split mirror compressed volumes (MBps)

Conclusion
In this evaluation of the Storwize V7000 V6.4 real-time compression feature with IBM i
database workload, an average compression savings of 66% was achieved. This
compression ratio of 1:3 can be considered representative because it also aligns well with the
typical IBM i tape drive compaction ratio.
From a performance perspective for the tested configuration and workload, a 23% increase in
IBM i database disk throughput with even lower response times was observed when V7000
compressed volumes were used versus non-compressed volumes. This improvement was
gained for a V7000/SAN Volume Controller disk backend, which was already highly utilized
when the non-compressed volumes were used.
A significant performance improvement by using real-time compression can be expected only

for already highly utilized disk backends and is achieved because of the compression, in
which less data must be read from and written to the disk backend. However, the reduction of
the disk backend utilization comes with a higher V7000/SAN Volume Controller CPU
utilization. This is due to the processing power that is required for the software real-time
compression engine so that for a proper V7000/SAN Volume Controller, sizing the aspect of
CPU utilization is not neglected.
2.5 Tips for implementing IBM Storwize V7000 in a IBM i

environment
This section describes the following useful tips for implementing the IBM Storwize V7000 in a
IBM i environment:
򐂰 2.5.1, “Storwize V7000 configuration options” on page 50
򐂰 2.5.2, “Host attachment” on page 51
򐂰 2.5.3, “Multipath” on page 52
򐂰 2.5.4, “Zoning SAN switches” on page 55
򐂰 2.5.5, “Data layout” on page 57
򐂰 2.5.6, “Solid-State Drives” on page 59
򐂰 2.5.7, “Booting from SAN” on page 60
򐂰 2.5.8, “Sizing for performance” on page 60
򐂰 2.5.9, “Performance monitoring” on page 61
򐂰 2.5.10, “Copy services considerations” on page 61
򐂰 2.5.11, “Translation from 520-byte blocks to 512-byte blocks” on page 62
2.5.1 Storwize V7000 configuration options

Different hardware and RAID options are available for the Storwize V7000 and can be
validated in terms of performance by using the Disk Magic sizing tool. You can configure the
RAID level and array width according to the solution that you modeled in Disk Magic tool. It is
considered a best practice to follow the default configuration options as outlined in the
Storwize V7000 GUI configuration wizard.
The default configuration option in the Storwize V7000 GUI is RAID 5 with a default array
width of 7+P for SAS HDDs, RAID 6 for Nearline HDDs with a default array width of 10+P+Q,
and RAID 1 with a default array width of 2 for SSDs.

It is considered a best practice to create a dedicated storage pool for IBM i with enough
managed disks that are backed by enough spindles to handle the expected IBM i workload.
Modeling with the Disk Magic tool by using actual client performance data can be performed
to size the storage system properly.
.
Note: Disk Magic is a tool that is used to size IBM disk subsystems that are based on the
performance and objectives of the workloads that are hosted on the disk subsystem. IBM
Disk Magic is a product to be used only by IBM personnel or an IBM Business Partner.
For more information about Disk Magic sizing for IBM i, see Chapter 5, Sizing for Midrange
Storage, in the IBM Redbooks publication IBM i and Midrange External Storage, SG24-7668.
2.5.2 Host attachment

Figure 2-54 shows that an IBM i system only logs in to a Storwize V7000 node a single time
from an I/O port adapter on the IBM i LPAR.
NPIV connection (shown with

LUN owned by Controller 0)
• Each adapter has active path to
Controller 0 Controller 1
• No common components Controller 0

(no single point of failure)
• Each adapter has path to passive
controller
• V7K automatically fails the LUN
over to Controller 1 in event of
Controller 0 failure
• IBM i automatically senses the new
active paths to the LUN
VIOS VIOS
For Business
vfiber vfiber
Figure 2-54 NPIV connection
The same connection considerations apply when you are connecting by using the native
connection option without VIOS. The following best practices guidelines are useful:
򐂰 Isolate host connections from remote copy connections (Metro Mirror or Global Mirror)
where possible.
򐂰 Isolate other host connections from IBM i host connections on a host port basis.
򐂰 Always have symmetric pathing by connection type. That is, use the same number of
paths on all host adapters that are used by each connection type.
򐂰 Size the number of host adapters that are needed based on expected aggregate
maximum bandwidth and maximum IOPS (use the Disk Magic tool or other common
sizing methods that are based on actual or expected workload).

2.5.3 Multipath
Multipath provides greater resiliency for SAN-attached storage. IBM i supports up to eight
paths to each LUN. In addition to the availability considerations, laboratory performance
testing shows that two or more paths provide performance improvements when compared to
a single path. Typically, two paths to a LUN is the ideal balance of price and performance.
Disk Magic tool: The Disk Magic tool supports multipathing only over two paths.
You might want to consider more than two paths for workloads where there is a high wait time
or where high I/O rates are expected to LUNs.
Multipath for a LUN is achieved by connecting the LUN to two or more ports that belong to
different adapters in an IBM i partition, as shown in the following examples:
򐂰 With native connection, although it is not a requirement, it is considered a recommended
practice for availability reasons to have the ports for multipath be on different physical
adapters in IBM i.
򐂰 With VIOS NPIV connection, the virtual Fibre Channel adapters for multipath should be
evenly spread across the different VIOS.
򐂰 With VIOS vSCSI connection, the virtual SCSI adapters for multipath should be evenly
spread across the different VIOS.
Figure 2-55 shows a high-level view of multipath in different types of connections.
Multipath with natively connected Multipath with VIOS NPIV connected Multipath with VIOS Virtual SCSI
Storwize V7000 Storwize V7000 connected Storwize V7000
IBM I IBM I
Multipath Multipath
Virtual SCSi
Virtual FC connections
IBM I connections
Multipath
hdisk hdisk
Switch Switch Switch Switch Switch Switch
Figure 2-55 IBM i multipath options
IBM i multipath provides resiliency in case the hardware for one of the paths fails. It also
provides performance improvement as multipath uses I/O load balancing in a round-robin
mode among the paths.
Every LUN in Storwize V7000 uses one V7000 node as a preferred node. The I/O traffic to or
from the particular LUN normally goes through the preferred node. If that node fails, the I/O is
transferred to the remaining node.

With IBM i multipath, all the paths to a LUN through the preferred node are active and the
path through the non-preferred node is passive. Multipath employs the load balancing among
the paths to a LUN that go through the node, which is preferred for that LUN.
Multipath with native and VIOS NPIV connections

In native and VIOS NPIV connections, multipath is achieved by assigning the same LUN to
multiple physical or virtual FC adapters in IBM i. The same LUN is assigned to multiple
WWPNs, each from one port in a physical or virtual FC adapter, with each virtual FC adapter
assigned to different VIOS.
For clarity, this description is limited to a multipath with two WWPNs. By using the preferred
practice of switch zoning, you can achieve the situation where four paths are established from
a LUN to the IBM i where two of the paths going through adapter 1 (in NPIV also through
VIOS 1) and two of the paths going through adapter 2 (in NPIV also through VIOS 2). From
the two paths that go through each adapter, one goes through the preferred node and other
goes through the non-preferred node. Therefore, two of the four paths are active, each of
them going through different adapter, and different VIOS if NPIV is used. Two of the paths are
passive, each of them going through different adapter, and different VIOS if NPIV is used.
IBM i multipathing uses a round-robin algorithm to balance the I/O among the paths that are
active.
Figure 2-56 shows the detailed view of paths in multipath with native or VIOS NPIV
connection to the Storwize V7000. The solid lines refer to active paths, and the dotted lines
refer to the passive paths. The red lines present one switch zone and green lines present the
other switch zone.
Figure 2-56 Native or NPIV connection multipath view

Figure 2-57 show the IBM i view of paths to the LUN that is connected in VIOS NPIV. You can
see two active and two passive paths are established to each LUN.
Display Disk Path Status
Serial Resource Path

ASP Unit Number Type Model Name Status
1 1 YF033F00001D 2145 050 DMP001 Active
DMP007 Passive
DMP008 Active
DMP002 Passive
1 2 YF033F00001E 2145 050 DMP003 Passive
DMP010 Active
DMP005 Active
DMP009 Passive
1 3 YF033F00001F 2145 050 DMP004 Active
DMP012 Passive
DMP006 Passive
DMP011 Active
Press Enter to continue.
F3=Exit F5=Refresh F9=Display disk unit details

F11=Display encryption status F12=Cancel
Figure 2-57 Displaying disk path status
Multipath with a VIOS vSCSI connection

In this type of connection, the LUN in Storwize V7000 is assigned to multiple VIOS. IBM i
establishes one path to the LUN through each VIOS. For clarity, this description is limited to a
multipath with two VIOS.
The LUN reports as device hdisk in each VIOS. The I/O rate from VIOS (device hdisk) to the
LUN uses all the paths that are established from VIOS to Storwize V7000. Multipath across
these paths and load balancing and I/O through a preferred node are handled by the VIOS
multipath driver. The two hdisks that represent the LUN in each VIOS are mapped to IBM i
through different virtual SCSI adapters. Each of the hdisks reports in to IBM i as a different
path to the same LUN (disk unit). IBM i establishes multipath to the LUN by using both paths.
Both paths are active and the load balancing in a round-robin algorithm is used for the I/O
traffic.

Figure 2-58 shows the detailed view of paths in multipath with VIOS vSCSI connected to IBM
Storwize V7000.
Figure 2-58 VIOS vSCSI multipath
IBM i uses two paths to the same LUN, each path through one VIOS to the relevant hdisk
connected with a virtual SCSI adapter. Both paths are active and the IBM i load balancing
algorithm is used for I/O traffic. Each VIOS has eight connections to the Storwize V7000;
therefore, eight paths are established from each VIOS to the LUN. The I/O traffic through
these paths is handled by the VIOS multipath driver.
Figure 2-59 shows both paths for the LUN in IBM i.
Display Disk Path Status
Serial Resource Path

ASP Unit Number Type Model Name Status
1 1 YTXBA98C5SSL 6B22 050 DMP274 Active
DMP275 Active
Figure 2-59 Disk paths from IBM i client
2.5.4 Zoning SAN switches

With native connection and the connection with VIOS NPIV, it is considered a best practice to
zone the switches so that one WWPN of one IBM i port is in zone with two ports of Storwize
V7000, with each port from one node canister. This way, you ensure resiliency for the I/O to or
from a LUN assigned to that WWPN. If the preferred node for that LUN fails, the I/O rate
continues by using the non-preferred node.

Note: A port in a virtual FC adapter in IBM i has two WWPNs. For connecting external
storage, use the first WWPN, the second WWPN is used for Live Partition Mobility (LPM).
Therefore, it is a good idea to zone both WWPNs if you plan to use LPM, otherwise zone
just the first WWPN.
When connection with VIOS virtual SCSI is used, is recommend to zone one physical port in
VIOS with all available ports in Storwize V7000, or with as many ports as possible to allow
load balancing. There are a maximum of eight paths that are available from VIOS to the
Storwize V7000. Storwize V7000 ports zoned with one VIOS port can be evenly spread
between Storwize V7000 node canisters.
Examples of zoning for VIOS vSCSI connection

Figure 2-60 shows the use of one physical FC port in VIOS with a mapped IBM i virtual Fibre
Channel client adapter, which was zoned with 2 * ports in Storwize V7000, 1 * ports from
each node canister. In this way, the IBM i client LPAR see one active and one passive path for
each configured IBM i volume on the Storwize V7000.
Figure 2-60 Zoning with one VIOS adapter

Figure 2-61 shows the use of two adapters in VIOS, zone one port from each adapter with 4 *
V7000 ports, with two ports from each canister. In this way, the Storwize V7000 is balanced
between VIOS ports and between the Storwize V7000 node canisters and do not exceed the
eight paths from VIOS to Storwize V7000.
Figure 2-61 Zoning with two VIOS adapters
2.5.5 Data layout

Selecting an appropriate data layout strategy depends on your primary objectives.
Spreading workloads across all components maximizes the utilization of the hardware
components. This includes spreading workloads across all the available resources. However,
it is always possible when sharing resources that performance problems can arise because of
contention on these resources.
To protect critical workloads, you can isolate them to minimize the chance that non-critical
workloads can affect the performance of critical workloads.
A storage pool is a collection of managed disks from which volumes are created and
presented to the IBM i system as LUNs. The primary property of a storage pool is the extent
size, which is 1 GB by default with Storwize V7000 release 7.1 and 256 MB in earlier
versions. This extent size is the smallest unit of allocation from the pool.
When you add managed disks to a pool, they can have the following performance
characteristics:
򐂰 Same RAID level.
򐂰 Roughly the same number of drives per array.

򐂰 Same drive type (SAS, NL_SAS, and SSD unless Easy Tier® is used) because data from
each volume is spread across all MDisks in the pool so the volume performs
approximately at the speed of the slowest MDisk in the pool.
򐂰 The exception to this rule is that if Easy Tier is used, you can have two different tiers of
storage in the same pool, but the MDisks within the tiers can still have the same
performance characteristics.
Workload isolation is most easily accomplished where each ASP or LPAR has its own
managed storage pool. This configuration ensures that you can place data where you intend
with I/O activity balanced between the two nodes or controllers on the Storwize V7000.
Important: Make sure that you isolate critical workloads. A best practice is to have only
IBM i LUNs on any storage pool (rather than mixed with non-IBM i). If you mix production
and development workloads in storage pools, make sure that you understand this might
affect production performance.
LUN compared to disk arm

The Storwize V7000 LUN that is connected to an IBM i reports in to the IBM i as a disk unit.
IBM i storage management employs the management and performance functions as though
the LUN was a disk arm. In fact, the LUN often is spread across multiple physical disk arms in
the Storwize V7000 disk pool, where the LUN uses some capacity from each disk arm.
All of the disk arms in the disk pools are shared among all the LUNs that are defined in that
disk pool. Figure 2-62 shows an example of a Storwize V7000 disk pool with three disk arrays
of V7000 internal disk arms (MDisks) and a LUN that was created in the disk pool, with the
LUN using an extent from each disk array in turn.
Figure 2-62 Storwize V7000 disk pool
LUN size
The maximum supported LUN size by IBM i is 2 TB - 1 Byte; that is, 2 TB LUNs are not
supported by IBM i.
The number of LUNs that are defined often is related to the wait time component of the
response time. If there are insufficient LUNs, wait time often increases. The sizing process
determines the correct number of LUNs that are required to access the needed capacity,
which meets performance objectives.

The number of LUNs drives the requirement for more FC adapters on the IBM i because of
the addressing restrictions of IBM i if you are using native attachment. Each path to a LUN
counts towards the maximum addressable LUNs on each IBM i IOA.
For any ASP, define all the LUNs to be the same size where 80 GB is the recommended
minimum LUN size. A minimum of six LUNs for each ASP or LPAR is considered a best
practice.
To support future product enhancements, it is considered a best practice that load source
devices be created with at least 80 GB. A smaller number of larger LUNs reduces the number
of I/O ports that are required on the IBM i and the Storwize V7000. In an IASP environment,
you might use larger LUNs in the IASPs, but SYSBAS might require more, smaller LUNs to
maintain performance.
The Disk Magic tool does not always accurately predict the effective capacity of the ranks
depending on the DDM size that is selected and the number of spares that are assigned. The
IBM Capacity Magic tool can be used to verify capacity and space utilization plans.
To calculate the number of LUNs that you require, you can use a tool, such as the IBM
Systems Workload Estimator, which is found at the following website:
http://www-912.ibm.com/wle/EstimatorServlet
Adding LUNs to ASP

Adding a LUN to an ASP generates I/O activity on the rank as the LUN is formatted. If there is
production work that is sharing the same rank, you might see a performance effect. For this
reason, it is considered a best practice that you schedule adding LUNs to ASPs outside peak
intervals.
2.5.6 Solid-State Drives

One of the most exiting innovations to happen to enterprise storage is solid-state drives
(SSDs). Solid-state storage means using a memory-type device for mass storage rather than
the use of a spinning disk or tape.
Solid-state storage technology has the following benefits:

򐂰 Significantly improved performance for hard-to-tune, I/O bound applications. No code
changes are required.
򐂰 Reduced floor space. Can be filled near 100% without performance degradation.
򐂰 Greater IOPS.
򐂰 Faster access times.
򐂰 Reduced energy use.
The use of SSDs with the Storwize V7000 is through Easy Tier. Even if you do not plan to
install SSDs, you can still use Easy Tier to evaluate your workload and provide information
about the benefit you might gain by adding SSDs. Easy Tier is included with the Storwize
V7000. However, it does require you to purchase a license and obtain a license key on a
Storwize V3700.
When Easy Tier automated management is used, it is important to allow Easy Tier some
space to move data. Do not allocate 100% of the pool capacity, but leave some capacity
deallocated to allow for Easy Tier migrations.

At a minimum, leave one extent available per tier in each storage pool. However, for optimum
use of future functions, plan to leave 10 extents available total per pool.
There also is an option to create a disk pool of SSDs in Storwize V7000 and to create an
IBM i ASP that uses disk capacity from an SSD pool. The applications that are running in that
ASP can experience a performance boost.
Note: IBM i data relocation methods (such as ASP balancing and media preference) are
not available to use with SSD in Storwize V7000.
2.5.7 Booting from SAN

All connection options of native, VIOS NPIV, and VIOS virtual SCSI support boot from SAN.
The LoadSource is on a Storwize V7000 LUN, which is connected the same way as the other
LUNs. There are no special requirements for LoadSource connection.
When you are installing the IBM i operating system with disk storage on a Storwize V7000,
you can select from one of the available Storwize V7000 LUNs when the installation prompts
you to select the load source disk.
When you are migrating from internal disk drives or from another storage system to the
Storwize V7000, you can use IBM i ASP balancing to migrate all of the disk capacity, except
the LoadSource. After the non-LoadSource data is migrated to the Storwize V7000 with load
balancing, you can then migrate the LoadSource by using the DST function to copy disk unit
data to copy from the previous disk unit to the LUN in Storwize V7000. The Storwize V7000
LUN must be of equal or greater size than the disk unit that was previously used for the
LoadSource. This type of migration can be done with the Storwize V7000 connections of
native, VIOS NPIV, and VIOS vSCSI.
2.5.8 Sizing for performance

It is important to size a storage subsystem that is based on I/O activity rather than capacity
requirements alone. This is true of an IBM i environment because of the sensitivity to I/O
performance. IBM has excellent tools for modeling the expected performance of your
workload and configuration. This section provides some guidelines for sizing for performance.
However, these guidelines provide a starting point only for sizing with the appropriate tools.
LUN size is flexible. Choose the LUN size that gives you enough LUNs for good performance
according to the Disk Magic tool. A best practice is to start modeling at 80 GB.
It is equally important to ensure that the sizing requirements for your SAN configuration also
take into account the other resources that are required when Copy Services is enabled. Use
the Disk Magic tool to model the overheads of replication (Global Mirror and Metro Mirror),
particularly if you are planning to enable Metro Mirror.
It is also a good idea to perform a bandwidth sizing for Global Mirror and Metro Mirror.
Note: The Disk Magic tool does not support modeling FlashCopy® (which is also known as
Point-in-Time Copy functions). Make sure that you do not size the system to maximum
recommended utilizations if you want to also use FlashCopy snapshots for backups.

You must collect IBM i performance data to be analyzed. Generally, you collect performance
data for a week’s worth of performance for each system or LPAR and analyze the result
reports for sizing.
Each set of reports includes print files for the following reports:
򐂰 System report: Disk utilization (required)
򐂰 Component report: Disk activity (required)
򐂰 Resource Interval report: Disk utilization detail (required)
򐂰 System report: Storage pool utilization (optional)
2.5.9 Performance monitoring

After your storage subsystem is installed, it is essential that you continue to monitor the
performance of the subsystem. IBM i Performance Tools reports provide information about
I/O rates and response times to the server. This allows you to track trends in increased
workload and changes in response time. Review these trends to ensure that your storage
subsystem continues to meet your performance and capacity requirements. Make sure that
you are current on fixes to ensure that the Performance Tools reports are reporting your
external storage correctly.
If you have multiple servers that are attached to a storage subsystem, particularly if you have
other platforms that are attached in addition to IBM i, it is essential that you have a
performance tool that enables you to monitor the performance from the storage subsystem
perspective.
The IBM Tivoli® Storage Productivity Center license product provides a comprehensive tool
for managing the performance of Storwize V7000. Collect data from all attached storage
subsystems in 15-minute intervals. If there is a performance problem, IBM asks for this data.
There is a simple performance management reporting interface that is available through the
IBM Storwize V7000 GUI. This provides a subset of the performance metrics that are
available from the IBM Tivoli Storage Productivity Center tool.
For more information about the IBM Tivoli Storage Productivity Center tool, see this website:
http://www-03.ibm.com/software/products/us/en/tivostorprodcent/
2.5.10 Copy services considerations

The Storwize V7000 has two options for Global Mirror: the classic Global Mirror and the
Change Volumes enhancement, which allows for a flexible and configurable RPO with which
Global Mirror can be maintained during peak periods of bandwidth constraint.
Change Volumes are not currently supported by PowerHA, so it is essential that you size the
bandwidth to accommodate the peaks or otherwise risk an effect to production performance.
There is a limit of 256 Global Mirror with change volume relationships per system.
The current zoning guidelines for mirroring installations advise that a maximum of two ports
on each Storwize V7000 node canister be used for mirroring, with the remaining two ports on
the node not having any visibility to any other cluster. If you are experiencing performance
issues when mirroring is operating, implementing zoning in this fashion might help to alleviate
this situation.

Consulting services are available from the IBM STG Lab Services to assist in the planning
and implementation of Storwize V7000 Copy Services in an IBM i environment. For more
information, see this website:
http://www-03.ibm.com/systems/services/labservices/platforms/labservices_i.html
2.5.11 Translation from 520-byte blocks to 512-byte blocks

IBM i disks have a block size of 520 bytes. Most fixed block (FB) storage devices are
formatted with a block size of 512 bytes, so a translation or mapping is required to attach
these to IBM i.
Note: The IBM DS8000 Storage system supports IBM i with a native disk format of 520
bytes.
IBM i performs the following change of the data layout to support 512-byte blocks (sectors) in
external storage: for every page (8 * 520-byte sectors) it uses another ninth sector; it stores
the 8-byte headers of the 520-byte sectors in the ninth sector; therefore, it changes the
previous 8* 520-byte blocks to 9* 512-byte blocks. The data that was previously stored in 8 *
sectors is now spread across 9 * sectors, so the required disk capacity on Storwize V7000 is
9/8 of the IBM i usable capacity. Conversely, the usable capacity in IBM i is 8/9 of the
allocated capacity in Storwize V7000.
Therefore, when an Storwize V7000 is attached to IBM i (whether through vSCSI, NPIV, or
native attachment), this mapping of 520:512 byte blocks means that you have a capacity
overhead and lose around 11% of the configured SAN storage capacity, which is not available
for IBM i user data.
The effect of this translation to IBM i disk performance is negligible.

3
Chapter 3. Setting up VIOS

This chapter describes how to set up an IBM PowerVM Virtual I/O Server (VIOS) by using the
Hardware Management Console (HMC) graphical user interface (GUI), which was introduced
in HMC Version 7.7.7.0 Service Pack 1. This chapter also explains how to set up VIOS by
using a command-line interface (CLI).
The configuration of vSCSI and N-Port ID Virtualization (NPIV) host attachment are also
explained by using both GUI and CLI interfaces, as the VIOS handles I/O requests to the
storage.
Finally, the network setup between the client partitions and the external network is described.
This explanation starts with a simple setup that uses one virtual network for the
communication between the partitions and the external network and ends with multiple virtual
networks to isolate traffic between the partitions, forwarding these VLAN tags to the external
network, and balancing the VIOS partitions for network load.

򐂰 Before you start
򐂰 VIOS setup by using HMC GUI
򐂰 VIOS setup by using the CLI
򐂰 Mapping storage through VIOS to IBM i
򐂰 Network setup
򐂰 Customizing
򐂰 Troubleshooting

3.1 Before you start
The two DVDs that are called “PowerVM Virtual I/O Server Installation Media” are required,
as physical media or as image files.
If your setup depends on multipath disk drive connections and vSCSI, the Subsystem Device
Driver Path Control Module (SDDPCM) driver is required. The SDDPCM driver provides for
specific SAN storage features, such as load balancing, that is not included in the native MPIO
multipath driver.
The SDDPCM driver can be downloaded from this website:

http://www-01.ibm.com/support/docview.wss?uid=ssg1S4000201
The following steps are not explained in detail. For more information about how to create a
logical partition and assign resources to it, see the IBM Redbooks publication IBM Power
Systems HMC Implementation and Usage Guide, SG24-7491:
1. Create the logical partition from the HMC. Select the IBM Power server that hosts the
VIOS from the HMC GUI on the left.
2. Click Configuration  Create Partition  VIO Server and assign the necessary
resources for storage and network.
3. Assign a minimum set of physical adapters to the VIOS partitions for presenting the
storage to the client partitions as needed. Add Ethernet adapters for VIOS management
and Ethernet virtualization.
The following minimum partition requirements must be met:

򐂰 One physical Ethernet adapter for external network access
򐂰 One physical storage adapter (for VIOS disk and client virtualization)
򐂰 Two virtual server serial adapters for console connections
򐂰 One or two virtual server SCSI adapters (one for virtual optical device; one for vSCSI
disks)
򐂰 One or two virtual server Fibre Channel adapters (one for NPIV disks; one for external
NPIV Fibre Channel tape device)

Table 3-1 lists the VIOS design limits for IBM i.
Table 3-1 VIOS design limits for IBM i

Design Item Maximum limit
Number of virtual target SCSI devices per virtual SCSI connection 16
Number of virtual devices per virtual NPIV port 64
Number of storage volumes per virtual Fibre Channel connection (NPIV) 64
Number of client LPARs (hosts) per physical Fibre Channel port (NPIV) 64
Number of NPIV virtual WWPN pairs per Power System 32000
Number of physical volumes (LUNs) per volume group (storage pool) 1024
Number of logical volumes per volume group (storage pool) 1024
Number of virtual adapters per LPAR 65535
Number of virtual Ethernet adapters per LPAR 32767
Number of 802.1Q VLAN tags per Shared Ethernet Adapter (SEA) 20
3.2 VIOS setup by using HMC GUI

Complete the following steps to set up your VIOS partition by using the HMC GUI:
Important: When it comes to network setup, make sure that communication with the HMC
is possible by using the specified IP addresses. This is used for the setup of the partitions
and is necessary to communicate with the partitions after the setup to add resources
dynamically.
1. On the HMC GUI, activate the VIOS partition by selecting the partition and clicking
Activate  Profile, as shown in Figure 3-1.
Figure 3-1 Activating the VIOS partition
Chapter 3. Setting up VIOS 65

2. In the Activate Logical Partition window (see Figure 3-2), select Yes to install the Virtual
I/O Server as part of activation process. Click OK to start the installation wizard by using
the HMC GUI.
Figure 3-2 Starting the VIOS installation wizard
The setup wizard starts the partition and scans for network hardware.
Important: Make sure that no console is connected to this partition because this
causes an error in hardware detection.
3. There are three possible ways to select a source for the installation media. A good
approach for IBM i clients is to use the physical media, which must be inserted into the
HMC optical drive. As an alternative, image files can be imported, which are on a remote
FTP site; for example, on an existing IBM i partition.
If physical media is used for the installation, continue to step 8 on page 68.

4. To import the image file into the local repository, select the Virtual I/O Server installation
source Local Repository and click Import, as shown in Figure 3-3.
Figure 3-3 Installing VIOS from a local repository
5. In the Import Virtual I/O Server Image window (see Figure 3-4), enter all of the necessary
information and click OK.
Figure 3-4 Importing the VIOS images
6. Figure 3-5 shows the loading panel. Wait until the files are uploaded to the HMC and then
continue with the installation.
Figure 3-5 Importing process

Attention: Do not use blanks or special characters for the repository name.
Tip: It is possible to manage the Virtual I/O Server Image Repository from the left side
of the HMC in the HMC Management menu.
7. Select the newly created repository to begin the installation process. Verify that the
network information is correct and then continue with step 10 on page 68.
8. As shown in Figure 3-6, all of the necessary information to install the VIOS partition from
physical media in the HMC optical drive was entered. Select the Virtual I/O Server
installation source of DVD and specify the correct network adapter and all the network
details for your VIOS partition. Click OK.
Figure 3-6 Installing VIOS from DVD
9. After all of the information is copied from the first installation media, a window opens, in
which you are prompted for the second media to be inserted into the HMC optical drive.
Exchange the DVD and click OK to continue the installation.
10.You see the message that is shown in Figure 3-7 after the installation is complete.
Figure 3-7 Installation success message

3.3 VIOS setup by using the CLI
Before you set up the VIOS by using the CLI, ensure that you created the VIOS partition
profile as described in 3.1, “Before you start” on page 64.
Important: Make sure that the VIOS installation media is inserted into the IBM Power
System optical drive and that the adapter that is connected to the drive is assigned to the
VIOS partition.
Complete the following steps:

1. On the HMC GUI, activate the VIOS partition by selecting the partition and clicking
Activate  Profile, as shown in Figure 3-8.
Figure 3-8 Activating the VIOS partition
2. In the Activate Logical Partition window (see Figure 3-9), click Advanced. Select SMS and
click OK to change the boot mode. Activate the VIOS partition by clicking OK.
Figure 3-9 Starting the VIOS partition without starting the installation wizard
3. Complete the following steps to open a console connection to the VIOS:

a. Connect by using Secure Shell (SSH) to the HMC.
b. Log on with hscroot or a similar account.
c. Run the vtmenu command to open a server or partition selection.

Tip: The console window can be closed by entering ~.
d. Browse to the VIOS partition and select it, as shown in Example 3-1.
Example 3-1 Opening a console window

[email protected]:~> vtmenu
Retrieving name of managed system(s) . . . z1436be:

----------------------------------------------------------
Partitions On Managed System: fz1436be:
OS/400 Partitions not listed
----------------------------------------------------------
1) "z1436cvios3:" Running
Enter Number of Running Partition (q to quit): 1
Opening Virtual Terminal On Partition z1436cvios3. . .
Open in progress..
4. Complete the following steps to select the boot device:

a. Select Select Boot Options and press Enter.
b. Select Select Install/Boot Device and press Enter.
c. Select Select 1st Boot Device and press Enter.
d. Select CD/DVD and press Enter.
e. Select the media type that corresponds to the optical device and press Enter.
f. Select the device number that corresponds to the optical device and press Enter.
g. Set the boot sequence to configure the first boot device. The optical device is now the
first device in the Current Boot Sequence list.
h. Exit the SMS menu by pressing X and confirming that you want to exit SMS.
5. Complete the following steps to install the VIOS:
a. Select the console that you want and press Enter.
b. Select a language for the Base Operating System (BOS) menus and press Enter.
c. Select Start Install Now with Default Settings and press Enter.
d. Select Change/Show Installation Settings and Install to change the installation
and system settings. This is necessary if more than one disk for VIOS is mapped to the
system or for more advanced virtualization features are used, such as LPM.
Figure 3-10 on page 71 shows the restrictions on each PowerVM edition.

Figure 3-10 IBM PowerVM editions
e. Select Continue with Install and press Enter. The system reboots after the
installation is complete.
6. Finish the installation by configuring an IP address to the VIOS partition, as shown in
Example 3-2.
Example 3-2 Configuring an IP address

$ lsdev en*
$ lsdev -dev ent0 -vpd
ent0 U78AA.001.WZSJG7K-P1-C7-T1 4-Port Gigabit Ethernet
PCI-Express Adapter (e414571614102004)
PCIe2 4-port 1GbE Adapter:

FRU Number..................00E2873
EC Level....................D77470
Customer Card ID Number.....576F
Part Number.................00E2872
Feature Code/Marketing ID...5899
Serial Number...............YL502032N09E
Manufacture ID..............6CAE8B01EF14
Network Address.............6CAE8B01EF14
ROM Level.(alterable).......10050160
Hardware Location Code......U78AA.001.WZSJG7K-P1-C7-T1
PLATFORM SPECIFIC
Name: ethernet
Node: ethernet@0
Device Type: network
Physical Location: U78AA.001.WZSJG7K-P1-C7-T1
$ cfgassist
VIOS TCP/IP Configuration
* Hostname [z1436cvios3]
* Internet ADDRESS (dotted decimal) [9.5.216.252]

Network MASK (dotted decimal) [255.255.255.0]
* Network INTERFACE en0
Default Gateway (dotted decimal) [9.5.216.1]
NAMESERVER
Internet ADDRESS (dotted decimal) [9.10.255.200]
DOMAIN Name [rch.stglabs.ibm.com]
CableType bnc +
3.4 Mapping storage through VIOS to IBM i

The VIOS partitions are the base to provide storage, whether internal or external to the IBM i
client partitions. If internal storage is used, only vSCSI connections are possible.
The limitations of vSCSI and NPIV are listed in Table 3-1 on page 65. If the focus of the setup
is for performance, it is considered a best practice to use NPIV instead of vSCSI.
3.4.1 Setting up vSCSI connections

The management of vSCSI disks is built into the HMC GUI. It is also possible to configure
everything from the VIOS command line. For multipath environments, the installation of the
IBM SDDPCM provides load balancing between the available Fibre Channel paths.
Downloading and installing the SDDPCM multipath driver

Complete the following steps to download and install the SDDPCM multipath driver:
1. Check which level of AIX® is the base for the installed VIOS. This can be done by using
the oem_platform_level command, as shown in Example 3-3.
The SDDPCM driver can be downloaded from this website:
Example 3-3 SDDPCM installation
$ oem_platform_level
AIX Version:
6.1.0.0
$ tar xf devices.sddpcm.61.rte.tar
$ cfgassist install_update
Install Software
* INPUT device / directory for software .

* SOFTWARE to install [_all_latest] +
PREVIEW only? (install operation will NOT occur) no +
COMMIT software updates? yes +
SAVE replaced files? no +
AUTOMATICALLY install requisite software? yes +
EXTEND file systems if space needed? yes +
OVERWRITE same or newer versions? no +
VERIFY install and check file sizes? no +
Include corresponding LANGUAGE filesets? yes +
DETAILED output? no +
Process multiple volumes? yes +

ACCEPT new license agreements? no +
[MORE...8]
$ shutdown -restart
2. Transfer the downloaded files to the VIOS partition by using FTP. Run the commands that
are shown in Example 3-3. They include a change of environment from VIOS commands
to a broadened selection of AIX commands to finding the AIX version. Then, return to the
VIOS environment, extract the downloaded archive, open the wizard to install the
SDDPCM drivers, and restart the VIOS.
Attention: If IBM i partitions are running and no secondary VIOS partition is configured
for failover, a restart of the VIOS partition does affect them all.
3. Test if the SDDPCM driver is working by switching to the OEM setup environment by using
the oem_setup_env and pcmpath query device commands.
Mapping disks from VIOS to IBM i partition by using HMC GUI

To map disks from VIOS to an IBM i partition, complete the following steps:
1. From the HMC GUI, click Systems Management  Serversand select the Power
System.
2. From the Tasks menu, click Configuration  Virtual Resources  Virtual Storage
Management. Select the VIOS partition that you want to work with from the drop-down
menu and then click Query.
3. The result is displayed as shown in Figure 3-11. In the first tab (Virtual Disks), virtual disks
can be created and mapped to the client partitions. This creates files in the VIOS that
serve as disk units, which are comparable to mounting storage by using a Network Server
Description (NWSD).
In the third tab (Physical Volumes), physical volumes can be mapped from the VIOS to the
client partition. With the use of external storage, such as the IBM Storwize Family, the disk
sizing can be done in the SAN before they are mapped to the VIOS. Select the disk to be
connected and click Modify assignment.
Figure 3-11 VIOS virtual storage management

4. In the Modify Physical Disk Partition Assignment window, select the partition and adapter
ID, as shown in Figure 3-12. Click OK. Disks can be added to partitions when they are
active, but can be removed only from partitions that are inactive. The Assigned Partition
column in Figure 3-11 changes to the partition name and slot ID that is selected.
Figure 3-12 Selecting partition and slot ID for disk
The disk is now visible within IBM i and can be added to the ASP as normal.
Mapping disks from VIOS to IBM i partition using CLI

To map disks by using the VIOS CLI, complete the following steps:
1. Connect to the VIOS partition by using an SSH utility. Log on as padmin user.
2. Use the lsdev, lsmap, and mkvdev or rmvdev commands to display your disk devices, list
the active mappings, and add or remove disks to or from partitions.
Example 3-4 shows how to display all available disks by using the hdisk command.
Tip: The location code V1-C6 that is shown in Example 3-5 references the VIOS
partition ID 1 (V1) and virtual adapter ID 6 (C6).
Example 3-4 Displaying all available disks

$ lsmap | grep hdisk
hdisk0 Available MPIO FC 2145
hdisk1 Available MPIO FC 2145

In Example 3-5 the active mappings are displayed and the hdisk1 device is mapped to the
IBM i partition.
Example 3-5 Displaying disk mappings and adding a disk to the host
$ lsmap -all
SVSA Physloc Client Partition

ID
--------------- -------------------------------------------- ------------------
vhost2 U8205.E6C.06F759R-V1-C6 0x00000004
VTD NO VIRTUAL TARGET DEVICE FOUND
$ mkvdev -vdev hdisk1 -vadapter vhost2

vtscsi0 Available
$ lsmap -all

ID
--------------- -------------------------------------------- ------------------
vhost2 U8205.E6C.06F759R-V1-C6 0x00000004
VTD vtscsi0
Status Available
LUN 0x8100000000000000
Backing device hdisk1
Physloc
U78AA.001.WZSJ2AV-P1-C2-T1-W500507680210DD40-L1000000000000
Mirrored false

3. In the IBM i service tools, the LUN ID from VIOS is visible as a physical location for the
logical resource, as shown in Figure 3-13.
You can select this display by clicking STRSST  Start a service tool  Hardware
service manager  Logical hardware resources  System bus resources 
Resources associated with IOP before the correct 290A adapter  Display detail at the
disk unit.
Auxiliary Storage Hardware Resource Detail
Description . . . . . . . . . . . . : Disk Unit

Type-model . . . . . . . . . . . . : 6B22-050
Status . . . . . . . . . . . . . . : Operational
Serial number . . . . . . . . . . . : YWAD6K386TL6
Part number . . . . . . . . . . . . :
Resource name . . . . . . . . . . . : DPH001
Physical location . . . . . . . . . : U8205.E6C.06F759R-V4-C3-T1-
L8100000000000000
Figure 3-13 Location of logical hardware resource
3.4.2 Mapping storage by using NPIV

Figure 3-14 shows four IBM i partitions that are hosted by a dual VIOS configuration. Each
IBM i partition uses two virtual Fibre Channel (FC) adapters to get redundant storage paths
through each VIOS partition.
Figure 3-14 N-Port ID virtualization
For the virtual client adapters, the HMC generates two World Wide Port Numbers (WWPNs).
The first WWPN is used for the SAN zoning. The second WWPN is important for Live
Partition Mobility (LPM) and is used to create the partition profile at the remote system.
The server-client pairs can be mapped to the same physical HBA and are visible to the SAN
and to the physical HBA WWPNs. Often, only the virtual WWPNs are used for further SAN
zoning operations because they are connected to the IBM i client partition.

To map disks by using the HMC GUI, complete the following steps:
1. From the HMC GUI, click Systems Management  Servers and then select the Power
Systems that you want to use.
2. From the Tasks menu, click Configuration  Virtual Resources  Virtual Storage
Management. Select the VIOS partition that you want to work with from the drop-down
menu and click Query.
3. The result is shown in Figure 3-15. In the last tab (Virtual Fibre Channel) of this window,
virtual server Fibre Channel adapters can be mapped to physical ports. The location of the
physical ports is shown in the Physical Location Code column. Distribute the client
partitions across the physical adapters according to the I/O load.
Select the Fibre Channel port to work with and then click Modify partition connections.
Figure 3-15 VIOS virtual Fibre Channel management
4. Figure 3-16 shows the active connections for the host bus adapter port fcs0. The
connections can be modified by selecting and clearing the check boxes. Click OK to save
the changes.
Figure 3-16 Modifying virtual Fibre Channel assignments
All disks that are zoned to the IBM i partition and connected to the IBM i virtual WWPN from
the IBM Storwize Family SAN storage are now visible within IBM i and can be added to the
ASP as normal. If disks are added or removed from the SAN, the changes are immediately
visible to the IBM i partition without further configuration of the VIOS.

Mapping disks from VIOS to IBM i partition by using the CLI
To map disks by using the VIOS CLI, complete the following steps:
1. Connect to the VIOS partition by using an SSH utility and log on as the padmin user.
2. Use the lsnports, lsmap, and vfcmap commands to display your host bus adapter devices,
list the active mappings, and add or remove virtual connections to or from the partitions.
Example 3-6 shows displaying all available adapters and the location codes. The output
contains the following fields:
– Physical port name (name)
– Physical port location code (physloc)
– Fabric support (fabric)
– Total number of NPIV ports (tports)
– Available number of NPIV ports (aports)
– Supported number of WWPNs (swwpns)
– Available number of WWPNs (awwpns)
Example 3-6 Displaying host bus adapters and location codes

$ lsnports
name physloc fabric tports aports swwpns awwpns
fcs0 U78AA.001.WZSJ2AV-P1-C2-T1 1 64 62 2048 2040
In Example 3-7, the active mappings are displayed and the virtual server host bus adapter
is mapped to the IBM i partition. The physical location code of the adapter is set to V1-C7
because it is a virtual adapter with the adapter ID 7 on the VIOS partition.
Example 3-7 Displaying disk mappings and adding a virtual server port to the host bus adapter
$lsmap -all -npiv
Name Physloc ClntID ClntName ClntOS

------------- ---------------------------------- ------ -------------- -------
vfchost2 U8205.E6C.06F759R-V1-C7 4
Status:NOT_LOGGED_IN
FC name: FC loc code:
Ports logged in:0
Flags:1<NOT_MAPPED,NOT_CONNECTED>
VFC client name: VFC client DRC:
$ vfcmap -vadapter vfchost2 -fcp fcs0

$ lsmap -all -npiv
Name Physloc ClntID ClntName ClntOS
------------- ---------------------------------- ------ -------------- -------
vfchost2 U8205.E6C.06F759R-V1-C7 4 z1436ca IBM i
Status:LOGGED_IN
FC name:fcs0 FC loc code:U78AA.001.WZSJ2AV-P1-C2-T1
Ports logged in:1
Flags:a<LOGGED_IN,STRIP_MERGE>
VFC client name:DC02 VFC client DRC:U8205.E6C.06F759R-V4-C7

3. In the IBM i service tools, the LUN ID from VIOS is visible as a physical location for the
logical resource, as shown in Figure 3-17.
You can select this display by clicking STRSST  Start a service tool  Hardware
service manager  Logical hardware resources  System bus resources 
Resources associated with IOP before the correct 6B25 adapter  Display detail at the
disk unit.
The disk type equals the type of the storage. The serial number of the disk contains the
LUN ID, which is visible in the IBM Storwize Family web-based interface.
Auxiliary Storage Hardware Resource Detail
Description . . . . . . . . . . . . : Disk Unit

Type-model . . . . . . . . . . . . : 2145-050
Status . . . . . . . . . . . . . . : Operational
Serial number . . . . . . . . . . . : YF033F000066
Part number . . . . . . . . . . . . :
Resource name . . . . . . . . . . . : DMP274
Physical location . . . . . . . . . : U8205.E6C.06F759R-V4-C12-T1-
W500507680210DD40-L0
Figure 3-17 Location of logical hardware resource
3.5 Network setup

Network setup in VIOS can be a complex topic. It is considered a best practice to set up at
least two VIOS instances for failover reasons. This way, VIOS updates can be done
concurrently and the network infrastructure can be used redundantly.
The network setup in VIOS consists of the following steps:

򐂰 3.5.1, “Optional step: LACP setup for adapter redundancy in each VIOS” on page 80
򐂰 3.5.2, “Creating VLAN-capable trunk Ethernet adapters” on page 80
򐂰 3.5.3, “Shared Ethernet Adapter setup” on page 83 for external network access and failure
resiliency
򐂰 3.5.4, “Transporting VLANs to the external network” on page 85 to Transport VLAN tags
out of the IBM Power System
Figure 3-18 on page 80 displays the complete VIOS networking setup, including all of these
steps.

Power System
IBM i LPAR 1 IBM i LPAR 2 IBM i LPAR 3
Virtual Virtual Virtual Virtual

Ethernet Ethernet Ethernet Ethernet
VID = 2500 VID = 2500 VID = 3983 VID = 3984
POWER Hypervisor
Control Channel VID=42
ha_mode=sharing
Trunk Adapter A Trunk Adapter B Trunk Adapter C Trunk Adapter D
VID = 2500, 3983 VID = 3984 VID = 2500, 3983 VID = 3984
VIOS1 Priority = 1 Priority = 1 Priority = 2 Priority = 2 VIOS2
(Primary) (Backup)
Shared Ethernet Adapter (SEA) Shared Ethernet Adapter (SEA)

Priority = 1 Priority = 2
Link Aggregation 802.3ad Link Aggregation 802.3ad

ent ent ent ent
(1Gb) (1Gb) (1Gb) (1Gb)
Ethernet Network
Inactive Trunk Adapter
Figure 3-18 VIOS networking
3.5.1 Optional step: LACP setup for adapter redundancy in each VIOS
The link aggregation control protocol (LACP) allows for Ethernet port redundancy. This
depends on the correct switch configuration. It is also a good idea to coordinate this
configuration with the network department.
A wizard guides you through the setup of LACP. Example 3-8 shows the LACP configuration
for ent0 and ent1.
Example 3-8 LACP configuration for ent0 and ent1

$ mkvdev -lnagg ent0,ent1
ent14 Available
en14
et14
3.5.2 Creating VLAN-capable trunk Ethernet adapters

As shown in Figure 3-18, a virtual trunk Ethernet adapter is required on the VIOS partitions to
become part of the SEA configuration that was described in 3.5.3, “Shared Ethernet Adapter
setup” on page 83.
Complete the following steps to set up the Ethernet adapter by using the HMC GUI:
1. From the HMC GUI, select Systems Management  Servers  IBM Power System.
2. Select the primary VIOS partition.
3. In the Tasks area, click Configuration  Manage Profiles  profile name.

4. Click the Virtual Adapters tab to create the virtual adapters for this partition, as shown in
Figure 3-19. Click Actions  Create Virtual Adapter  Ethernet Adapter.
Figure 3-19 VIOS virtual adapter configuration
5. In this step, a virtual trunk Ethernet adapter is created that can bridge traffic to the external
network and allow for multiple VLANs, as shown in Figure 3-20 on page 82.
If you do not plan on having multiple virtual networks, do not select the IEEE 802.1q
compatible adapter option, which adds the VLAN tagging options.
Tip: It is considered a best practice to use only one default VLAN for client traffic and
another internal VLAN for the shared Ethernet adapter high availability.
Tip: VLAN IDs can range 0 - 4095. The IDs 0 and 4095 are reserved. VLAN ID 1 is the
default VLAN ID for most hardware providers. Do not use these IDs for VLAN setups to
avoid unexpected behavior.
It is important that the VLAN ID that is added to the packets that are leaving the client
partitions (the default VLAN ID number) is not removed on entering the Virtual I/O Server.
This is what happens when a default configuration is used.
It is for this reason that the default VLAN ID on the Virtual I/O Servers virtual Ethernet
adapter (as shown in Figure 3-20 on page 82) must be set to an unused VLAN ID (in this
example, 42 is used) that is not used outside the system.
If a packet arrives with this VLAN ID (42), the VLAN ID tag is stripped off (untagged) and
cannot be forwarded to the correct external VLAN. If this was sent through the shared
Ethernet adapter to the external physical network, it arrives at the Ethernet switch as
untagged.
Click OK to save the settings and close the window.

Figure 3-20 Creating a virtual Ethernet trunk adapter for external network access
6. Repeat steps 3 on page 80 through step 5 on page 81 for the secondary VIOS. Use a
priority value of two for the virtual trunk Ethernet adapter.
7. Add another virtual Ethernet adapter (as shown in Figure 3-21) for the control channel of
the SEA with as high a VLAN ID as possible. Make sure to coordinate with the network
department. When the SEA is created, it checks for a virtual Ethernet adapter, with no
access to the external network, and starts at the highest VLAN ID.
Tip: With an active RMC connection between VIOS and HMC, all changes to an active
partition can be done by using DLPAR operations. Take care to save all changes to the
LPAR profile afterward.
Figure 3-21 Creating a control channel virtual Ethernet adapter

3.5.3 Shared Ethernet Adapter setup
The Shared Ethernet Adapter (SEA) bridges the traffic between the virtual trunk Ethernet
adapter and the physical adapter interface. It is also used for high availability reasons.
Complete the following steps to create an SEA adapter:
1. From the HMC GUI, click Systems Management  Servers  IBM Power System.
2. From the Tasks area, click Configuration  Virtual Network Management.
3. As shown in Figure 3-22, select the default VSwitch and select the VLAN ID of 42, as this
was the defined virtual trunk Ethernet adapter that was created in the example in step 5 on
page 81. Click Create SEA.
Tip: You can select each VLAN ID to get a list of the associated adapter resource name
in every LPAR.
Figure 3-22 HMC virtual network editor

4. Select your primary and secondary VIOS partitions and the physical network interface, as
Figure 3-23 SEA creation
5. The wizard returns to the Virtual Network Management window. If you select the VLAN ID
of 4094, you can see that this VLAN acts now as the control channel VLAN for the newly
created SEA, as shown in Figure 3-24.
Figure 3-24 HMC virtual network editor with control channel VLAN
The SEA configuration is now finished and client partitions with virtual Ethernet adapters
on VLAN 42 can access the external network. VLAN tags are not yet exported to the
external network. If you decide to use other VLAN tags in VLAN 2500 and 3983, continue
with 3.5.4, “Transporting VLANs to the external network” on page 85.

6. This optional step adds a virtual trunk Ethernet adapter to the SEA and reconfigures it to
use load sharing, as shown in Figure 3-18 on page 80. For more information, see
Example 3-9.
Important: The two steps that are shown in Example 3-9 must be run in each VIOS
partition, starting with the primary site.
Example 3-9 Adding a trunk adapter to the SEA and change ha_mode attribute
ent6: SEA
ent2: trunk adapter which is already part of the SEA
ent5: new trunk adapter adding to the SEA
$ chdev -dev ent6 -attr virt_adapters=ent2,ent5

ent6 changed
Changing the SEA to Load Sharing mode.
$ chdev -dev ent6 -attr ha_mode=sharing
ent6 changed
3.5.4 Transporting VLANs to the external network

Complete the following steps to transport the VLAN IDs (that can be used only within the
vSwitch until now) to the external network. The network hardware that is attached must be
configured for VLAN tagging:
1. Connect to the VIOS partitions by using SSH or open a console connection.
2. Use the mkvdev command to add VLAN devices to the shared Ethernet adapter, as shown
in Example 3-10.
Example 3-10 Exporting the VLAN IDs using the SEA adapter ent6
$ mkvdev -vlan ent6 -tagid 2500
ent7 Available
en7
et7
ent8 Available
en8
et8
ent9 Available
en9
et9
3.6 Customizing
This section describes how to ease navigation within the VIOS shell. This section also shows
how to set up the SNMP daemon to monitor VIOS performance remotely. The following topics
are covered:
򐂰 3.6.1, “Changing ksh to bash style” on page 86
򐂰 3.6.2, “Configuring monitoring by using SNMP” on page 86
򐂰 3.6.3, “Creating an image repository for client D-IPL” on page 87

3.6.1 Changing ksh to bash style
To permanently change the VIOS shell to a more intuitive way of operation, follow the
commands that are shown in Example 3-11. This example uses the editor vi that is
preinstalled with VIOS.
Use the o key to insert a new line and change to editing mode. Use the Esc key to exit editing
mode.
Outside the editing mode, you can use commands to browse vi. Enter :d to delete a line and
:x to exit and save the file.
Example 3-11 Changing ksh to bash style

$ oem_setup_env
# chmod 644 /usr/ios/cli/.profile
# vi /usr/ios/cli/.profile
add the following lines to the file:
export ENV=/usr/ios/cli/.kshrc
export PS1="$(ioscli hostname)$ "
alias __A=ècho "\020"` #ctrl-p => up
alias __B=ècho "\016"` #ctrl-n => down
alias __C=ècho "\006"` #ctrl-f => right
alias __D=ècho "\002"` #ctrl-b => left
alias __H=ècho "\001"` # Home
alias __F=ècho "\005"` # End
set -o emacs
exit the editor via ESC and :x
# exit
3.6.2 Configuring monitoring by using SNMP

Complete the following steps to enable the Simple Network Management Protocol (SNMP)
daemon:
1. Edit the file snmpdv3.conf with the vi editor, as described in 3.6.1, “Changing ksh to bash
style” on page 86.
2. Edit the file rc.tcpip to enable the SNMP daemon auto start.
3. Restart the SNMP daemons.

For more information, see Example 3-12.
Example 3-12 Configuring SNMP daemons

$ oem_setup_env
# vi /etc/snmpdv3.conf
Put a # before the following line to comment it out:

VACM_VIEW defaultView 1.3.6.1.4.1.2.6.191 - excluded -
Add the next four lines to the end of the file, save, and exit:
VACM_GROUP group1 SNMPV2c public -
VACM_ACCESS group1 - - noAuthNoPriv SNMPv2c defaultView - defaultView -
COMMUNITY public public noAuthNoPriv 0.0.0.0 0.0.0.0 -
VACM_VIEW defaultView 1 - included -
# vi /etc/rc.tcpip
Uncomment the following lines, save, and exit:

start /usr/sbin/snmpd "$src_running"
start /usr/sbin/hostmibd "$src_running"
start /usr/sbin/snmpmibd "$src_running"
start /usr/sbin/aixmibd "$src_running"
Restart the services:

# for s in snmpd snmpmibd hostmibd aixmibd; do stopsrc -s $s; sleep 2; done
# for s in snmpd snmpmibd hostmibd aixmibd; do startsrc -s $s; sleep 2; done
# exit
3.6.3 Creating an image repository for client D-IPL

If no physical DVD drive is available to the client partition, create an image repository that is
on the VIOS partitions disk. This can be done by using the HMC GUI or from the VIOS CLI.
Attention: Unlike image catalogs in IBM i, the images in the repository are not autoloaded.
Before you start, a virtual DVD drive must be created on the VIOS partition by using the VIOS
command line. Complete the following steps:
1. Create a virtual server SCSI adapter on the VIOS partition profile and a matching virtual
client SCSI adapter on the client partition that is used as alternative Load Source device.
2. Run the mkvdev command from the VIOS CLI for the created device (as shown in
Example 3-13 on page 88) to create a virtual optical device.
Tip: Use the -dev parameter with the mkvdev command to set a device name, such as
vtopt_iLPAR1 for later reference.

Example 3-13 Creating a virtual optical device
$ lsmap -all
ID
--------------- -------------------------------------------- ------------------
vhost0 U8205.E6C.06F759R-V1-C4 0x00000003
VTD NO VIRTUAL TARGET DEVICE FOUND
$ mkvdev -fbo -vadapter vhost0

vtopt0 Available
$ lsmap -all
ID
--------------- -------------------------------------------- ------------------
vhost0 U8205.E6C.06F759R-V1-C4 0x00000003
VTD vtopt0
Status Available
LUN 0x8100000000000000
Backing device
Physloc
Mirrored N/A
Creating a repository and adding images by using the HMC GUI

Complete the following steps to create the repository, add the I_BASE_01 image, and assign
it to the IBM i client partition:
1. Select your IBM Power Server and open the Virtual Storage Management from the Tasks
area by clicking Configuration  Virtual Resources  Virtual Storage Management.
2. Select the VIOS partition and click Query.
3. As shown in Figure 3-25 on page 89, click Create Library in the Optical Devices tab to
add a new library.

Figure 3-25 Managing optical devices
4. Figure 3-26 shows the Create Media Library window. Enter the media library size. The
maximum size depends on the VIOS disk that is used. Ensure that it is large enough to
store the image files. Click OK.
Figure 3-26 Creating the media library
5. Transfer the image file by using FTP or an SCP client to the VIOS server.

6. Add the transferred image file to the repository, as shown in Figure 3-27. Enter a media
name by which the image can be identified and the full path to the image file. Click OK to
create a Virtual Media object.
Figure 3-27 Adding virtual media
Figure 3-28 shows the upload progress.
Tip: You can delete the image file from the VIOS after it is copied from its location to the
repository.
Figure 3-28 Creating the virtual optical media

7. Select the image file and assign it to the virtual optical device, as shown in Figure 3-29.
Figure 3-29 Assigning the virtual optical media
8. Select the partition and virtual optical driveSelect the Force reassignment on running
partitions option and click OK, as shown in Figure 3-30. Use read-only as the access
mode to load the image into multiple virtual drives.
Figure 3-30 Assigning the virtual media

Creating a repository and adding images by using the VIOS CLI
Example 3-14 shows how to create an image repository, add an image to the repository, and
load the image for the client partition by using the VIOS CLI. Use read-only as the access
mode to load the image into multiple virtual drives.
Example 3-14 Creating a repository, adding an image, and loading the image
$ lssp
Pool Size(mb) Free(mb) Alloc Size(mb) BDs Type
rootvg 51136 27008 64 0 LVPOOL
$ mkrep -sp rootvg -size 26G
Virtual Media Repository Created
Repository created within "VMLibrary" logical volume
$ mkvopt -name IBASE71G -file I_BASE_01_710G.iso -ro
$ loadopt -disk IBASE71G -vtd vtopt0
$ lsrep
Size(mb) Free(mb) Parent Pool Parent Size Parent Free
26515 24134 rootvg 51136 384
Name File Size Optical

Access
IBASE71G 2381 vtopt0 ro
$ unloadopt -vtd vtopt0
3.7 Troubleshooting
This section describes how to improve VIOS performance and find bottlenecks. This section
also shows how to restart the Resource Monitoring and Control (RMC) connection and which
information is necessary for a support call and how to gather that information.
The following topics are covered:

򐂰 3.7.1, “Measuring performance” on page 92
򐂰 3.7.2, “Restarting the RMC connection” on page 95
򐂰 3.7.3, “Collecting troubleshooting data for support” on page 96
3.7.1 Measuring performance

Starting with VIOS 2.2.2, a tool to measure the performance and check the VIOS
configuration is available. Complete the following steps:
1. Run the part command (as shown in Example 3-15) to collect the performance
information.
Example 3-15 Gathering performance statistics

$ part
usage: part {-i INTERVAL | -f FILENAME} [-t LEVEL] [--help|-?]
-i <minutes> interval can range between 10-60
-f <file> any nmon recording
-t <level> 1 - Basic logging, 2 - Detailed logging
-? usage message
$ part -i 30
part: Reports are successfully generated in z1436cvios3_131011_20_26_48.tar

2. Download the archive from the VIOS and extract it at your workstation.
3. Open the vios_advisor_report.xml file with a web browser.
The resulting output is split into different sections. In Figure 3-31, the overview of the
system is displayed. Click the icons to see an explanation for each item.
Figure 3-31 System configuration
Figure 3-32 shows information about average and maximum I/O activity within the
measured interval.
Figure 3-32 VIOS I/O and network activity

Figure 3-33 shows the details about the physical disk drives that are visible to the VIOS.
This example uses NPIV attached disks. Only the disk that contains the VIOS is shown.
Figure 3-33 VIOS disk drives
Figure 3-34 displays the average CPU usage. The VIOS footprint is small if you use NPIV.
The CPU load increases with a higher virtualized network load.
Figure 3-34 VIOS CPU usage
Figure 3-35 shows the memory usage. If the VIOS starts to page, increase the size of the
memory by using a DLPAR operation without the need for downtime.
Figure 3-35 VIOS memory usage

As shown in Figure 3-35 on page 94, there is a warning about the memory usage. Click
the warning icon to see a detailed explanation, as shown in Figure 3-36. As this system
still has about 20% of its memory unused, it is not necessary to take action, but the
amount of available memory can decrease soon.
Figure 3-36 Memory warning explanation
Alternatively, the VIOS Advisor tool can be downloaded from this website:
https://www.ibm.com/developerworks/mydeveloperworks/wikis/home/wiki/Power%20System
s/page/VIOS%20Advisor
The tool allows for a longer collection period of up to 1440 minutes and provides greater detail
on I/O activity and network configuration.
3.7.2 Restarting the RMC connection

When the connection between HMC and VIOS is interrupted, the RMC connection can be
lost. Without an active RMC connection, dynamic LPAR operations are no longer possible.
Use the commands that are shown in Example 3-16 to check the connection state and restart
the connection.
Example 3-16 Checking and restarting the RMC connection

$ oem_setup_env
# /usr/sbin/rsct/bin/rmcdomainstatus -s ctrmc
# /usr/sbin/rsct/install/bin/recfgct
0513-071 The ctcas Subsystem has been added.
0513-071 The ctrmc Subsystem has been added.
0513-059 The ctrmc Subsystem has been started. Subsystem PID is 9502788.
# /usr/sbin/rsct/bin/rmcdomainstatus -s ctrmc
Management Domain Status: Management Control Points
I A 0x69e3763d06415474 0002 9.5.103.98
I A 0x48864326026f84b4 0001 9.5.104.205
# exit
The rmcdomainstatus command now shows a connection to both attached HMCs.

3.7.3 Collecting troubleshooting data for support
This section describes the information you need to collect for support.
Gather information about the type, model, and serial number of the IBM Power System.
Use the commands that are shown in Example 3-17 to collect information about the installed
operating system levels.
Example 3-17 Collecting operating system information

RZMB02VIOS1$ ioslevel
2.2.2.1
RZMB02VIOS1$ oem_platform_level
AIX Version:
6.1.0.0
To collect information about the virtual adapter configuration (including the error logs), run the
snap command and download the resulting snap.pax.Z file by using FTP to your workstation.
The process resembles what is shown in Example 3-18 and runs for several minutes.
Example 3-18 Running the snap command

$ snap
Gathering data. This may take 5 to 10 minutes. Please wait.
Checking space requirement for general information.............. done.

Checking space requirement for tcpip information................ done.
Checking space requirement for kernel information............... done.
Checking space requirement for printer information.............. done.
Checking space requirement for dump information..................done.
Checking space requirement for sna information.................. done.
Checking space requirement for filesys information.............. done.
Checking space requirement for async information................ done.
Checking space requirement for lang information................. done.
Checking space requirement for XS25 information..................done.
Log directory set to /tmp/ibmsupt/ctsupt
Gathering information..............................................
Starting tar/compress process......
*******done*******
Checking space requirement for install information.............. done.
Checking space requirement for ssa information.................. done.
...
...
...
Gathering pcixscsi system information........................... done.
Gathering sissas system information............................. done.
Gathering IB system information..............................
Creating compressed pax file...
Starting pax/compress process... Please wait... done.
-rw------- 1 0 0 43131685 Oct 11 11:31 snap.pax.Z

Finished gathering data. Check your home directory for snap.pax.Z
$

4
Chapter 4. PowerHA and the IBM Storwize

Family
Increasing business demands for application availability require more clients of any size to
look for a solution that can help eliminate planned and unplanned downtimes for their IT
services.
An unplanned outage that in duration or recovery time that exceeds business expectations
can have severe implications, including unexpected loss of reputation, client loyalty, and
revenue. Clients who did not effectively plan for the risk of an unplanned outage, never
completed their installation of a high availability (HA) solution, or did not have a tested tape
recovery plan in place, are especially exposed to negative business effects.
This chapter describes various usage scenarios for IBM PowerHA SystemMirror for i with the
SAN Volume Controller and IBM Storwize storage servers. This chapter also describes the
possibilities, position them against each other, and gives you an overview of how to configure
them.

򐂰 Usage scenarios for IBM PowerHA SystemMirror for i
򐂰 Setting up IBM PowerHA SystemMirror for i

4.1 Usage scenarios for IBM PowerHA SystemMirror for i
The IBM PowerHA SystemMirror for i solution offers a complete end-to-end integrated
clustering solution for high availability (HA) and disaster recovery (DR). PowerHA provides a
data and application resiliency solution that is an integrated extension of the IBM i operating
system and storage management architecture. It also features the design objective of
providing application high availability through planned and unplanned outages.
Because data resiliency is managed within the IBM i storage management architecture, there
is no operator involvement. Just as there is no operator involvement with RAID 5 or disk
mirroring.
As shown in Figure 4-1, the architecture of PowerHA SystemMirror for i provides several
different implementation possibilities.
PowerHA SystemMirror for i (5770-HAS)
Cross Site Mirroring (XSM)

“HA Switchable Resources” - IBM i option 41
End-to-End Solution
Storage Command Line Interface

Cluster Resources in IBM i
Metro Mirror Global Mirror LUN Level Switching FlashCopy Geographic Mirroring
• SAN hardware • SAN • Single copy of IASP • SAN • IBM i storage mgmt
replication hardware switched between hardware page level replication
• Synchronous replication LPARs or server replication • Sync. or asynchronous
• IBM DS8000 or • Asynchronous • IBM DS8000 or • Point in time • Internal or ext. storage
Storwize family/ • IBM DS8000 or Storwize family • IBM DS8000 or • Supports direct, VIOS
Storwize family Storwize family & IBM i hosted storage
Figure 4-1 Overview of IBM PowerHA SystemMirror for i
Geographic mirroring offers IBM i clients an IBM i-based page-level replication solution for
implementing high availability and disaster recovery with any kind of IBM i-supported internal
or external storage solution. With IBM System Storage DS8000, SAN Volume Controller, or
IBM Storwize family, storage servers clients can use storage-based remote replication
functions for high availability and disaster recovery, LUN-level switching for local high
availability, and FlashCopy for reducing save window outages by enabling the creation of a
copy that is attached to a separate partition for offline backup to tape.
Which of these solutions make most sense in your environment depends largely on the
storage infrastructure in place.
The following solutions are described in this section:

򐂰 4.1.1, “Metro Mirror” on page 99
򐂰 4.1.2, “Global Mirror” on page 99
򐂰 4.1.3, “LUN level switching” on page 100
򐂰 4.1.4, “FlashCopy” on page 102
For more information about the Storwize Copy Services functions, see the IBM Redbooks
publication, IBM System Storage SAN Volume Controller and Storwize V7000 Replication
Family Services, SG24-7574.

4.1.1 Metro Mirror
Metro Mirror is a synchronous remote copy relationship between two volumes (VDisks) of
equal (virtual) size. When the remote copy relationship is established, the preferred primary
volume is designated as the master volume and the preferred secondary volume as the
auxiliary volume. If the secondary volume is available, every host write I/O that is sent to the
Metro Mirror primary volume is acknowledged back to the host only after it is committed to the
write cache of the primary Storwize family system and the secondary Storwize family system.
Figure 4-2 shows a typical setup with PowerHA and Metro Mirror. Each system has its own
system ASP that contains the operating system, license programs, and some other objects.
The IASP that contains application programs and data is mirrored from the primary storage
system to the secondary storage system. Data is kept synchronous on a page level.
Metro Mirror is commonly used with PowerHA to achieve a high availability and disaster
recovery solution for metropolitan distances up to 300 km when Storwize V3700, V5000,
V7000, or SAN Volume Controller is used.
SysBas
Metro Mirror of IASP SysBas
iASP iASP‘
Figure 4-2 Metro Mirror
4.1.2 Global Mirror

Global Mirror is an asynchronous remote copy relationship between two volumes (VDisks) of
equal (virtual) size. When the remote copy relationship is established, the preferred primary
volume is designated as the master volume and the preferred secondary volume as the
auxiliary volume. Every host write I/O sent to the Global Mirror primary volume is
acknowledged back to the host after it is committed to the write cache of both nodes for the
corresponding I/O group of the primary Storwize family system. At some later time (that is,
asynchronously), this write update is sent by the primary Storwize family system to the
secondary Storwize family system. Therefore, Global Mirror provides the capability to perform
remote copy over long distances, up to the maximum supported round-trip latency of 80 ms,
which exceeds the performance-related limitations of synchronous remote copy without host
write I/O performance affects that are caused by remote replication delays.
Chapter 4. PowerHA and the IBM Storwize Family 99

Though the data is sent asynchronously from the primary to the secondary Storwize family
system, the write ordering is maintained by sequence numbers that are assigned to
acknowledged host write I/Os and with the secondary applying writes in order by their
sequence number.
Consistency of the data at the remote site is maintained always. However, during a failure
condition, the data at the remote site might be missing recent updates that were not sent or
that were in-flight when a replication failure occurred. Therefore, the use of journaling to allow
for proper crash consistent data recovery is of key importance. It is also considered a best
practice to ensure crash consistent data recovery even when remote replication is not used.
This allows you to recover from a loss of transient modified data in IBM i main storage after a
potential sudden server crash due to a severe failure condition.
A log file is used by the Storwize family for Global Mirror to maintain write ordering. It also is
used to help prevent host write I/O performance impacts when the host writes to a disk sector
that is being transmitted or because of bandwidth limits that are still waiting to be transmitted
to the remote site. The Storwize family also uses shared sequence numbers to aggregate
multiple concurrent (and dependent) write I/Os to minimize its Global Mirror processing
overhead.
Global Mirror is commonly used with PowerHA to achieve disaster recovery where you want
to put some distance between your sites. You want to combine it with a local high availability
solution because the asynchronous implementation of Global Mirror does not ensure that the
last transaction reached the backup system.
4.1.3 LUN level switching

LUN level switching, as shown in Figure 4-3 on page 101, provides support for an automated
IBM i high availability solution by using a single copy of an independent ASP that can be
switched (or ASPs), which can be switched between two IBM i cluster nodes for a local high
availability solution.
It is supported for NPIV attachment and for native attachment of the IBM Storwize family. It is
also supported for an SAN Volume Controller split cluster environment, which can make it an
attractive solution, especially in heterogeneous environments where SAN Volume Controller
split cluster is used as the basis for a cross-platform, two-site, high-availability solution.
PowerHA
Sys iASP Sys

Bas Bas
SVC Split Cluster
SVC Quorum 3 SVC Quorum 2 SVC Quorum 1
Figure 4-3 LUN level switching
SAN Volume Controller split cluster

In a standard SAN Volume Controller configuration, all nodes physically are within the same
rack. Beginning with version 5.1, support was provided for split cluster configurations where
nodes within an I/O group can be physically separated from one another by up to 10
kilometers. This capability allows nodes to be placed in separate failure domains, which
provide protection against failures that affect a single failure domain.
The initial support for split cluster that was delivered in version 5.1 contained the restriction
that all communication between the SAN Volume Controller node ports cannot traverse
Inter-Switch Links (ISLs). This limited the maximum supported distance between failure
domains. Starting with SAN Volume Controller version 6.3, the ISL restriction was removed,
which allowed the distance between failure domains to be extended to 300 kilometers.
Additionally, in SAN Volume Controller version 6.3, the maximum supported distance for
non-ISL configurations was extended to 40 kilometers.
The SAN Volume Controller split cluster configuration provides a continuous availability
platform whereby host access is maintained if any single failure domain is lost. This
availability is accomplished through the inherent active architecture of the SAN Volume
Controller with the use of volume mirroring. During a failure, the SAN Volume Controller
nodes and associated mirror copy of the data remain online and available to service all host
I/O.
The split cluster configuration uses the SAN Volume Controller volume mirroring function.
Volume mirroring allows the creation of one volume with two copies of MDisk extents. The two
data copies, if placed in different MDisk groups, allow volume mirroring to eliminate the effect
to volume availability if one or more MDisks fails. The resynchronization between both copies
is incremental and is started by the SAN Volume Controller automatically. A mirrored volume
has the same functions and behavior as a standard volume. In the SAN Volume Controller
software stack, volume mirroring is below the cache and copy services. Therefore,
FlashCopy, Metro Mirror, and Global Mirror have no awareness that a volume is mirrored. All
operations that can be run on non-mirrored volumes can also be run on mirrored volumes.

LUN level switching is commonly used with an SAN Volume Controller split cluster setup. This
setup provides storage high availability through the SAN Volume Controller setup and server
high availability through PowerHA. Another common setup is a three node cluster where local
high availability is achieved by LUN level switching and disaster recovery is possible on the
third node in the cluster by using Metro or Global Mirror. In this three-node setup, you often
use an SAN Volume Controller split cluster on the primary site and a Storwize V7000 or
V3700 at the secondary site. Within the Storwize family, remote copy operations, such as
Global Mirror and Metro Mirror, are possible between the different members of the family.
Important: Although SAN Volume Controller split cluster is a supported environment with
PowerHA, there is no automatic change of the SAN Volume Controller preferred node
when a failover or switchover is done. Because of the latency that is involved when you are
reading from the remote SAN Volume Controller node, make sure that the distance
between the SAN Volume Controller nodes is close enough to meet your disk response
time expectations. In addition, you must have separate Fibre Channel adapters for the
system ASP and the IASP. Also, ensure that you set the preferred node correctly when you
are creating the vDisks on the SAN Volume Controller because changing this setting later
on requires the attached IBM i LPAR to be powered down.
4.1.4 FlashCopy
The IBM Storwize family FlashCopy function provides the capability to perform a point-in-time
copy of one or more volumes (VDisks). In contrast to the remote copy functions of Metro
Mirror and Global Mirror, which are intended primarily for disaster recovery and
high-availability purposes, FlashCopy typically is used for online backup or creating a clone of
a system or IASP for development, testing, reporting, or data mining purposes.
FlashCopy also is supported across heterogeneous storage systems that are attached to the
IBM Storwize family, but only within the same Storwize family system. Up to 4096 FlashCopy
relationships are supported per IBM Storwize family system, and up to 256 copies are
supported per FlashCopy source volume. With IBM Storwize family version 6.2 and later, a
FlashCopy target volume can also be a non-active remote copy primary volume, which eases
restores from a previous FlashCopy in a remote copy environment by using the FlashCopy
reverse function.
FlashCopy is commonly used to minimize downtimes for backup operations to tape. Instead
of saving data from the production system, you create a FlashCopy of that data and attach it
to another system where the actual save to tape is done, as shown in Figure 4-4 on page 103.
Production LPAR LPAR for Save
FlashCopy
IASP IASP
Figure 4-4 FlashCopy
In addition, FlashCopy can be used to provide a kind of disaster recovery option when large
changes are made to your application. Create the FlashCopy before you start the application
change. When you find that something does not work as expected, you can revert to the state
before the application changes without waiting for a lengthy restore operation.
For more information about how an IASP is suspended for a brief period, the FlashCopy is
taken, and the BRMS setup is done, see Chapter 6, “BRMS and FlashCopy” on page 173.
4.2 Setting up IBM PowerHA SystemMirror for i

This section describes the step-by-step setup instructions for configuring IBM PowerHA
SystemMirror for i with the IBM Storwize family. The section includes the following topics:
򐂰 4.2.1, “Requirements for PowerHA SystemMirror for i with SAN Volume Controller or
Storwize” on page 103
򐂰 4.2.2, “SAN Volume Controller split cluster and LUN level switching” on page 105
򐂰 4.2.3, “LUN level switching in combination with Global Mirror” on page 137
򐂰 4.2.4, “Metro Mirror setup” on page 150
򐂰 4.2.5, “Adding LUNs to an existing IASP” on page 150
4.2.1 Requirements for PowerHA SystemMirror for i with SAN Volume

Controller or Storwize
To work properly, PowerHA SystemMirror for i must communicate with the IBM Storwize
Family system. This is done by using Secure Shell (SSH). You must create an SSH key pair
on IBM i and attach the SSH public key to a user profile on the IBM Storwize Family system.
The corresponding private key file to be used is specified in the creation of the ASP copy
descriptions. It must be distributed to the same directory on all nodes in the cluster.
Generating the SSH key requires the following IBM i products to be installed:
򐂰 5770-SS1 option 33 (Portable Application Solution Environment)
򐂰 5733-SC1 base and option 1 (IBM Portable Utilities for i and OpenSSH, OpenSSL, zlib)

Note: The Enterprise edition of PowerHA is required for two-site Copy Services solutions.
Generating the key file is done on IBM i from Qshell. Example 4-1shows the commands in
Qshell to generate the key and the files that are created. Ensure that the directory you specify
in the keygen command exists before the command is issued. Also, the user profile you are
using when you are generating the key can have a maximum length of eight characters only.
Example 4-1 Generation of the SSH key pair in Qshell

> cd /QIBM/UserData/HASM/hads/.ssh/
$
> ssh-keygen -t rsa -f id_rsa -N ''
Generating public/private rsa key pair.
Your identification has been saved in id_rsa.
Your public key has been saved in id_rsa.pub.
The key fingerprint is:
76:49:e1:9e:04:0a:c5:e2:68:3a:d0:6b:0b:4b:e2:2e
[email protected]
$
> ls -la
total 64
drwxrwsrwx 2 powerha 0 8192 Oct 07 17:02 .
drwx--Sr-x 3 qsys 0 8192 Oct 07 15:19 ..
-rw------- 1 powerha 0 1679 Oct 07 17:02 id_rsa
-rw-r--r-- 1 powerha 0 414 Oct 07 21 17:02 id_rsa.pub
You then must import the id_rsa.pub file into a user on the Storwize family system. This user
must have a minimum role of Copy Operator to perform the functions that are needed for
PowerHA to work properly. If you want to use LUN level switching, make sure that the user
has a minimum role of Administrator because PowerHA must change host attachments when
switching the IASP to the secondary system.
Transfer the file to your PC and import it to the Storwize user, as shown in Figure 4-5.
Figure 4-5 Importing SSH public key file to SAN Volume Controller user
Make sure to distribute the id_rsa file to all nodes in your cluster to the same directory. The
user profile QHAUSRPRF must have at least *R data authority to the key file on each system.
4.2.2 SAN Volume Controller split cluster and LUN level switching
This section provides an example of configuring SAN Volume Controller split cluster and LUN
level switching. In this example, it is assumed that SAN Volume Controller split cluster was set
up correctly. Also, ensure that SAN zoning is set up so that the IASP can be reached from the
primary cluster node and from the secondary cluster node.
Complete the following steps by using the GUI. These steps cover creating the cluster,
creating the cluster resource group, and creating the ASP copy descriptions.
To complete these steps by using a CLI, see “Using CLI commands to setup SAN Volume
Controller split cluster and LUN level switching” on page 136:
1. Log in to the IBM Systems Director Navigator for i by using the following URL:
http://your_server_IP_address:2001

2. From the list of possible tasks that is shown in Figure 4-6, select PowerHA to access the
PowerHA configuration and management GUI.
Figure 4-6 Cluster creation from IBM Systems Director Navigator
3. As shown in Figure 4-7, there is no configured cluster. Click Create a new Cluster to start
the cluster creation wizard.
Figure 4-7 Creating a cluster
4. On the Create Cluster welcome panel, click Next to continue to the cluster setup wizard.
5. As shown in Figure 4-8, on the Create Cluster Name and Version panel, you must provide
a name for the cluster. With current fixes installed, the PowerHA version is 2.2. Click Next.
Figure 4-8 Cluster creation, specifying the cluster name and version

6. In the Create Cluster Local Node panel, provide information about the local cluster node,
as shown in Figure 4-9. Specify a node name and one or two IP addresses that are used
for cluster communication. You can enter these addresses manually or click the Use entry
from below drop-down menu and select the address from the list that is shown.
Tip: It is a common practice to use two different IP connections for the cluster heartbeat
so that if there is a partial network failure, cluster communication is still available.
Figure 4-9 Cluster creation, local node specifications
7. In the Create Cluster Additional Nodes panel, specify the information for the second node
in the cluster. Click Add, as shown in Figure 4-10, and specify the node name and the IP
addresses for the second cluster node. You can also specify whether clustering is started
after the creation of the cluster is finished.
Figure 4-10 Cluster creation, adding the second node

8. When all of the nodes in your cluster are specified, click Next, as shown in Figure 4-11.
Figure 4-11 Cluster creation, all nodes specified
9. In the Create Cluster Message Queue panel, you must decide whether you want to use a
cluster message queue, as shown in Figure 4-12. When a message queue is specified, a
message is sent to that message queue on the secondary system when there is a failure
on the primary system. You can also specify how long that message waits for an answer
and what the standard action is after that time passes. Click Next.
Figure 4-12 Cluster creation, cluster message queue

10.As shown in Figure 4-13, you see an overview of the configuration options that you chose.
Click Finish to start the cluster creation.
Figure 4-13 Cluster creation summary
11.Status messages appear as the cluster is created, as shown in Figure 4-14. Click Close to
return to the PowerHA main menu.
Figure 4-14 Cluster creation successful message
12.From the PowerHA main menu (as shown in Figure 4-15), select Cluster Nodes to edit
the cluster nodes that were created.
Figure 4-15 PowerHA main menu

13.Figure 4-16 shows the two nodes of the cluster that was created in this example. Currently,
there is no device domain that is specified for the two nodes. The device domain is needed
so that you can switch an IASP between the nodes.
Figure 4-16 Nodes in the cluster
14.Right-click each node in the cluster and select Properties, as shown in Figure 4-17.
Figure 4-17 Node properties selection
15.Figure 4-18 shows the properties for cluster node Z1436B23 from our example. There is
no entry for the device domain. Click Edit.
Figure 4-18 Cluster node properties

16.Define the name of the device domain, as shown in Figure 4-19. If there is already a
device domain that is created within the cluster, you can choose the device domain from
the drop-down menu.
Note: If an IASP exists on one of the cluster nodes, that node must be added first into
the device domain. Only the first node in a device domain can have an existing IASP.
Nodes with an existing IASP cannot be added to an existing device domain.
Figure 4-19 Specifying the device domain
17.Device domains must be defined for each node in the cluster. As shown in Figure 4-20, the
cluster node overview shows the active cluster nodes with their device domains.
Figure 4-20 Device domains added to each node in the cluster

18.Define the cluster resource group that is used to switch the IASP between the cluster
nodes. From the main PowerHA window, click Cluster Resource Group, as shown in
Figure 4-21.
Figure 4-21 PowerHA main menu, selecting Cluster Resource Groups
19.In the PowerHA Cluster Resource Groups window, select Create Cluster Resource
Group from the Select Action drop-down menu, as shown in Figure 4-22.
Figure 4-22 Specifying to create a cluster resource group
20.In the Create Cluster Resource Group Name and Type panel, specify a name for the
cluster resource group. The type must be Device, as shown in Figure 4-23, because you
want to switch an IASP device that uses this cluster resource group. Click Next.
Figure 4-23 Create Cluster Resource Group, specifying the name and type

21.In the Create Cluster Resource Group Recovery Domain panel, define the recovery
domain. The recovery domain defines which node in the cluster is the primary node and
the order of backup nodes in the cluster.
Add the primary node into the recovery domain as shown in Figure 4-24. A site name also
must be specified. Click Add to add the primary node into the recovery domain.
Figure 4-24 Create Cluster Resource Group, recovery domain for the primary node
22.Add the secondary node into the recovery domain. Specify Backup as the role. Ensure
that the site name you provide here is identical to the site name that you provided for the
primary node, as shown in Figure 4-25 on page 121. Click Add to add the node into the
recovery domain.
Figure 4-25 Create Cluster Resource Group, recovery domain for the backup node

23.The recovery domain overview shows the two cluster nodes with their node roles and site
names, as shown in Figure 4-26. Click Next.
Figure 4-26 Create Cluster Resource Group, recovery domains created
24.In the Create Cluster Resource Group Exit Program panel, you can define a cluster
resource group exit program. This program can be used to start functions that are specific
to your environment when failover or switchover is done; for example, to start a certain
application. For this example, an exit program is not used, as shown in Figure 4-27. Click
Next.
Figure 4-27 Create Cluster Resource Group, specifying an exit program

25.In the Create Cluster Resource Group Devices panel, define which device is controlled by
the cluster resource group. Specify the name of the IASP, as shown in Figure 4-28.
You also can specify whether the IASP is varied on at the secondary system after a
failover or switchover occurs. The server takeover IP address can be used to move one IP
address to the current primary node. Click Add to add the device into the cluster resource
group.
Figure 4-28 Create Cluster Resource Group, defining devices that are controlled by the cluster
26.You can add multiple devices into one cluster resource group. Click Next when the devices
you need are part of the cluster resource group, as shown in Figure 4-29.
Figure 4-29 Create Cluster Resource Group, device added

27.In the Create Cluster Resource Group Failover Message Queue panel, you can define a
failover message queue for the cluster resource group. As our example already defined a
cluster-wide failover message queue, it does not define another one at the cluster
resource group level, as shown in Figure 4-30. Click Next.
Figure 4-30 Create Cluster Resource Group, specifying a failover message queue
28.The Create Cluster Resource Group Summary panel that is shown in Figure 4-31 displays
the configuration options that were entered. Click Finish to create the cluster resource
group.
Figure 4-31 Create Cluster Resource Group, summary
The cluster resource group is created and is in an inactive status, as shown in Figure 4-32.
Figure 4-32 Cluster resource group created

29.The last step to set up LUN level switching with PowerHA SystemMirror for i is to create an
ASP Copy description. From the PowerHA main menu, click Independent ASPs, as
Figure 4-33 PowerHA main menu, selecting Independent ASPs
30.In the PowerHA Independent ASPs window, select ASP Sessions and Copy
Descriptions from the Select Action drop-down menu, as shown in Figure 4-34.
Figure 4-34 Selecting ASPs and Copy descriptions

31.In the ASP Sessions and Copy Descriptions window, select Add SVC ASP Copy
Description from the Select Action pull-down menu in the ASP Copy Descriptions
section, as shown in Figure 4-35. This option is used when the storage attached is a
member of the IBM Storwize family.
Figure 4-35 Adding an SAN Volume Controller ASP Copy description
32.In the Add SVC ASP Copy Description window, you must define the connection to the
SAN Volume Controller and the disk range and host connections from the SAN Volume
Controller setup for the LUNs of the IASP. As shown in Figure 4-36, enter the following
information:
– Enter a name for the ASP Copy Description.
– The ASP Device, the Cluster Resource Group, and the Site name must be the names
that were specified previously in our example.
– User Id is the user ID on the SAN Volume Controller that you imported the SSH key
into. This field is case-sensitive. Make sure that you enter the user ID exactly as you
specified it in the SAN Volume Controller.
– The Secure Shell Key File points to the SSH private key that is stored in the same
directory for all nodes in the cluster
– The IP Address is the network address of the SAN Volume Controller.
– Enter the disk range that is used for the IASP by clicking Add to add the information to
your configuration setup. This information can be found in the SAN Volume Controller
GUI. You can also add several disk ranges here. Make sure that the Use All Host
Identifiers option is selected.
Figure 4-36 SAN Volume Controller ASP Copy Description creation, adding a virtual disk range

33.You also need to enter the host identifier for all the nodes in the recovery domain. Click
Add Host Identifier for each node, enter the Host Identifier, and click Add to add the
information to your configuration, as shown in Figure 4-37. The host identifier can be
found in the SAN Volume Controller GUI.
Figure 4-37 SAN Volume Controller ASP Copy description creation, adding a host identifier for the primary node
34.Click Add to add the Recovery Domain Entry, as shown in Figure 4-38. Make sure that
you add the host identifier information for each node in the cluster.
Figure 4-38 SAN Volume Controller ASP Copy description creation, adding the primary node

35.Figure 4-39 shows the finished configuration with the disk range for the IASP and two
cluster nodes in the recovery domain and their corresponding host identifiers. Click OK to
finish the configuration.
Figure 4-39 SAN Volume Controller ASP Copy description creation, recovery domain completed
Figure 4-40 shows the completed setup. Click Back to Independent ASPS.
Figure 4-40 SAN Volume Controller ASP Copy description creation finished
36.Start the cluster resource group. From the main PowerHA menu, click Cluster Resource
Groups.

37.In the PowerHA Cluster Resource Groups window, right-click the cluster resource name
and select Start, as shown in Figure 4-41. The SAN Volume Controller with LUN level
switching by using PowerHA SystemMirror for i setup is now complete.
Figure 4-41 Starting the cluster resource group
Using CLI commands to setup SAN Volume Controller split cluster and
LUN level switching
You can use the CLI commands that are shown in Example 4-2 to complete this setup.
Example 4-2 CLI commands to set up the SAN Volume Controller split cluster and LUN level switching
On the backup system:
CRTDEVASP DEVD(IASP1) RSRCNAME(IASP1)
On the production system:

CRTCLU CLUSTER(SVC_CLU) NODE((Z1436B23 ('9.5.216.233' '192.168.10.23'))
(Z1436B22 ('9.5.216.230' '192.168.10.22')))
CLUMSGQ(QSYS/QSYSOPR) FLVWAITTIM(*NOMAX) FLVDFTACN(*CANCEL)
ADDDEVDMNE CLUSTER(SVC_CLU) DEVDMN(SVC_DMN) NODE(Z1436B23)
CRTCRG CLUSTER(SVC_CLU) CRG(SVC_CRG) CRGTYPE(*DEV)

EXITPGM(*NONE) USRPRF(*NONE)
RCYDMN((Z1436B23 *PRIMARY *LAST SITE1 *NONE)
(Z1436B22 *BACKUP *LAST SITE1 *NONE))
CFGOBJ((IASP1 *DEVD *ONLINE '10.10.10.10'))
ADDSVCCPYD ASPCPY(PROD) ASPDEV(IASP1) CRG(SVC_CRG) SITE(SITE1) NODE(*CRG)

SVCHOST(PowerHA '/QIBM/UserData/HASM/hads/.ssh/id_rsa' '9.5.216.80')
VRTDSKRNG((1 4 *ALL)) RCYDMN((Z1436B23 (3)) (Z1436B22 (4)))
STRCRG CLUSTER(SVC_CLU) CRG(SVC_CRG)
4.2.3 LUN level switching in combination with Global Mirror

LUN level switching can also be used with Global Mirror in a three-node cluster. In this setup,
LUN level switching provides local high availability. In addition, if the whole local site goes
down, Global Mirror provides a disaster recovery option to a remote site. Depending on the
distance between the two sites, you can also consider the use of Metro Mirror instead of
Global Mirror.
For more information about Metro Mirror, Global Mirror, and LUN level switching, see 4.1.1,
“Metro Mirror” on page 99, 4.1.2, “Global Mirror” on page 99, and 4.1.3, “LUN level switching”
on page 100.
It is assumed that the following steps were completed:

򐂰 The PowerHA setup for the local LUN level switching was completed, as described in
4.2.2, “SAN Volume Controller split cluster and LUN level switching” on page 105.
򐂰 Global Mirror for all volumes of the IASP were configured.
򐂰 The private SSH key that is used for communication was distributed between PowerHA
and the SAN Volume Controller to the third cluster node and the public SSH key was
assigned to a user on the secondary SAN Volume Controller. For more information, see
4.2.1, “Requirements for PowerHA SystemMirror for i with SAN Volume Controller or
Storwize” on page 103.
To complete these steps by using a CLI interface, see “Using CLI commands to setup LUN
level switching with Global Mirror in a three-node cluster” on page 149.
Complete the following steps by using the GUI interface:

1. Add the third cluster node into the existing cluster. To do this, select Cluster Nodes from
the PowerHA main menu in IBM Navigator for i.
2. In the PowerHA Cluster Nodes window, select Add Cluster Node from the Select Action
pull-down menu, as shown in Figure 4-42 on page 138.

Figure 4-42 Adding a cluster node to create a three-node cluster
3. In the Add Cluster Node window, enter the node name and the cluster IP addresses that
are used for cluster communication, as shown in Figure 4-43. Click OK.
Figure 4-43 Three-node cluster, specifying the third cluster node details
4. As shown in Figure 4-44, the cluster node is added and started. Click Close.
Figure 4-44 Three-node cluster: Third node added successfully
5. As shown in Figure 4-45, our example now has a three-node cluster. The third node was
not added to the device domain. You can add the node manually to the device domain or
PowerHA can handle this for you when the cluster node is added into the cluster resource
group. Click Back to PowerHA to return to the main PowerHA menu.
Figure 4-45 Three-node cluster: Node overview
6. In the PowerHA main menu, select Cluster Resource Groups.

7. In the PowerHA Cluster Resource Groups window, select Recovery Domain from the
Select Action pull-down menu, as shown in Figure 4-46.
Figure 4-46 Three-node cluster: Creating a recovery domain

8. In the PowerHA Cluster Resource Groups Recovery Domain window, select Add
Recovery Domain Node from the Select Action pull-down menu, as shown in
Figure 4-47.
Figure 4-47 Three-node cluster: Adding a recovery domain node
9. In the Add Recovery Domain Node window (as shown in Figure 4-48), enter the following
information:
– Node name of the third node.
– The node role is set to Backup with an order number of 2. In this example, there
already is a primary node and a first backup node that is defined in the cluster setup.
– Because this is a different site from the site that was used for LUN level switching, you
must specify a different site name here. Think of the site name as a representation for
each copy of the IASP in the recovery domain.
Click OK.
Figure 4-48 Three-node cluster: Specifying the recovery domain details
10.The new node is not yet added into the cluster device domain, as shown in the message
that is shown in Figure 4-49. Click OK and the new node is automatically added to the
same device domain that is used by the other nodes in the recovery domain.
Figure 4-49 Three-node cluster: Warning message

11.A status message appears as the node is added to the device domain, as shown in
Figure 4-50.
Figure 4-50 Three-node cluster: Status message
12.A status message is then displayed that shows that the third node was added to the
recovery domain, as shown in Figure 4-51. Click Close.
Figure 4-51 Three-node cluster: Node successfully added to recovery domain
13.The recovery domain now consists of three cluster nodes, as shown in Figure 4-52. Click
Back to Cluster Resource Groups and return to the main PowerHA menu.
Figure 4-52 Three-node cluster: Recovery domain overview
14.In the main PowerHA menu, select Independent ASP.

15.In the PowerHA Independent ASPs window, add an ASP Copy Description for the IASP on
the third node. Select ASP Sessions and Copy Descriptions from the Select Action
pull-down menu, as shown in Figure 4-53.
Figure 4-53 Three-node cluster: ASP session
16.In the PowerHA Independent ASPs, ASP Sessions and Copy Descriptions window, select
Add SVC ASP Copy Description from the Select Action pull-down menu (as shown in
Figure 4-54) because this example is working with an SAN Volume Controller/Storwize
V7000 environment.
Figure 4-54 Three-node cluster: Adding a SAN Volume Controller copy description

17.In the Add SAN Volume Controller ASP Copy Description window, specify the following
parameters for the ASP, as shown in Figure 4-55:
– ASP Copy Description name.
– ASP Device is the name of the IASP that is used.
– Name of the Cluster Resource Group.
– Enter the new Site Name that was created in step 9 on page 141.
– Define how PowerHA can talk to the storage system where the Global Mirror target
LUNs are located. The User Id is case-sensitive.
Click Add Virtual Disk Range.
Figure 4-55 Three-node cluster: Specifying copy description details
18.Enter the numbers for your virtual disk range from the SAN Volume Controller, as shown in
Figure 4-56. Click Add to add this information into the ASP copy description. You can
specify multiple disk ranges here. Click OK.
Figure 4-56 Three-node cluster: Adding the virtual disk range

19.Define an ASP session that tells PowerHA the relationship between ASP copy
descriptions. Select Start SVC ASP Session from the Select Action pull-down menu in
the ASP Sessions section, as shown in Figure 4-57.
Figure 4-57 Three-node cluster: Defining a SAN Volume Controller ASP session
20.In the Start SVC ASP Session window, specify the following information, as shown in
Figure 4-58:
– Type: Defines the relationship between the ASP copy descriptions. This example set
up a Global Mirror connection between the primary and the secondary LUNs.
– Specify the Cluster Resource Group that is used.
– You can also decide whether replication between the LUNs is reversed after doing a
switchover or a failover. By default, replication is not reversed after a failover so that the
original data in the primary IASP is preserved to provide a chance for error analysis.
Click OK.
Figure 4-58 Three-node cluster: Starting an SAN Volume Controller ASP session
The three-node cluster that uses LUN level switching for local high availability and Global
Mirror for remote disaster recovery is now ready for use.
Using CLI commands to setup LUN level switching with Global Mirror in
a three-node cluster
If you prefer to set up this environment by using CL commands, use the commands that are
shown in Example 4-3.
Example 4-3 Setup for adding Global Mirror by using CL commands

On the third cluster node:
On the primary cluster node:

ADDCLUNODE CLUSTER(SVC_CLU) NODE(Z1436C23 ('9.5.216.226' '192.168.10.123'))
ADDDEVDMNE CLUSTER(SVC_CLU) DEVDMN(SVC_DMN) NODE(Z1436C23)
ADDCRGNODE CLUSTER(SVC_CLU) CRG(SVC_CRG)

RCYDMN(Z1436C23 *BACKUP *LAST SITE2 *NONE)
ADDSVCCPYD ASPCPY(IASP_GM) ASPDEV(IASP1) CRG(SVC_CRG) SITE(SITE2) NODE(*CRG)

SVCHOST(powerha '/QIBM/UserData/HASM/hads/.ssh/id_rsa' '9.5.216.236')
VRTDSKRNG((1 4 *ALL))
STRSVCSSN SSN(GLOBALMIR) TYPE(*GLOBALMIR) CRG(SVC_CRG)

4.2.4 Metro Mirror setup
Setting up PowerHA SystemMirror for i in a Storwize family environment by using Metro Mirror
is comparable to the setup that was described in 4.2.3, “LUN level switching in combination
with Global Mirror” on page 137.
The main differences are the configuration on the Storwize system where you specify to use
Metro Mirror instead of Global Mirror. You must setup a cluster with two nodes in it. You then
create the cluster resource group. Make sure to specify different site names for the node
identifiers in the recovery domain. Create two SAN Volume Controller Copy Descriptions that
point to the LUNs that are used as Metro Mirror source volumes and Metro Mirror target
volumes. When you are starting the SAN Volume Controller session, make sure to specify
*METROMIR as the session type.
For more information about setup instructions for this environment, see PowerHA
SystemMirror for IBM i Cookbook, SG24-7994.
4.2.5 Adding LUNs to an existing IASP

Complete the following steps to add LUNs to an existing IASP that is used with PowerHA:
1. Define the new LUNs on the storage system and map them to the correct host.
2. If you are using Metro Mirror or Global Mirror, define remote copy relationships between
the LUNs on your storage system and start the copy relationship. When these copy
relationships reach a consistent status, you add the copy relationships into the existing
consistency group.
3. If you are using FlashCopy, make sure to add the new LUNs into the existing consistency
group that you are using.
4. Add the LUNs into the IASP on IBM i from SST.
5. Add the ranges of the newly created LUN ranges to the corresponding ASP copy
descriptions. You can use the Change SAN Volume Controller Copy Description
(CHGSVCCPYD) command or use the Navigator GUI by completing the following steps:
a. From Independent ASP link in the PowerHA GUI, select ASP Sessions and Copy
Description from the Select Action pull-down menu, as shown in Figure 4-59.
Figure 4-59 Adding LUNs to ASP session and copy description

b. In the PowerHA Independent ASP > ASP Sessions and Copy Descriptions window,
click the Properties for the IASP to which you added the LUNs, as shown in
Figure 4-60.
Figure 4-60 Adding LUNs, selecting properties of the IASP
c. In the Properties window, click Edit to change the ASP Copy Description, as shown in
Figure 4-61.
Figure 4-61 Adding LUNs, editing properties

d. In the Properties window, you can see the virtual disk range that is assigned to the
IASP. Click Add Virtual Disk Range, as shown in Figure 4-62.
Figure 4-62 Adding LUNs, adding virtual disk range
e. Enter the disk range of the newly added LUNs and click Add, as shown in Figure 4-63.
Figure 4-63 Adding LUNs, entering a new virtual disk range

f. When all of the new LUN ranges are entered, click Save (as shown in Figure 4-64) to
save the new configuration for the ASP copy description.
Figure 4-64 Adding LUNs, saving the new configuration
5
Chapter 5. Live Partition Mobility

By using Live Partition Mobility, you can migrate partitions and their hosted applications from
one physical server to another without disrupting the infrastructure services. The migration
operation, which takes a few seconds, maintains complete system transactional integrity. The
migration transfers the entire system environment, including processor state, memory,
attached virtual devices, and connected users.
This chapter describes the general steps that are part of a Live Partition Mobility operation. It
gives an overview of the prerequisites to use Live Partition Mobility and guides you through
the setup process.

򐂰 Usage scenarios for Live Partition Mobility
򐂰 License considerations with Live Partition Mobility
򐂰 Setting up Live Partition Mobility

5.1 Usage scenarios for Live Partition Mobility
IBM Power System servers are designed to offer the highest stand-alone availability in the
industry. Still, enterprises must occasionally restructure their infrastructure to meet new IT or
business requirements. By letting you move your running production applications from one
physical server to another, Live Partition Mobility allows for non-disruptive maintenance or
modification to a system. This mitigates the effect on partitions and applications that were
caused by the occasional need to shut down a system.
Because Live Partition Mobility works only when the current production system is up and
running, it is not a replacement for another high availability solution. Instead, it is another
feature that you want to use with your high availability solution.
Live Partition Mobility can help in reducing planned downtimes for hardware maintenance or
replacement of systems and in moving logical partitions to another system to achieve a more
balanced workload distribution. As the complete logical partition is moved to a new system
and nothing remains on the original system, Live Partition Mobility cannot help in scenarios
where you need access to one environment when you are performing productive work on
another environment. This includes operating system maintenance or application
maintenance.
The following steps provide an overview of the process that is involved in using Live Partition
Mobility to move a running LPAR to a new system:
1. As a starting point in Figure 5-1, the IBM i client partition is running on System #1. On
System #2, there is a Virtual I/O Server that is running that is connected to the storage
system that System #1 is using. As shown in Figure 5-1, there is no configuration that is
available on System #2 for the IBM i LPAR that we want to migrate to that system.
The environment is checked for required resources, such as main memory or processor
resources.
Power7 System #1 Power7 System #2
Suspended
IBM i Client
Partition
1
Check environment for
M M M M M M M M
required resources
A LIN1
DC01 CMN01
HMC
VLAN VLAN
Hypervisor Hypervisor
VASI vhost0 ent1 ent1 VASI
Mover ent2 en2 en2 ent2 Mover

Service vtscsi0 Service
SEA (if) (if) SEA
fcs0 ent0 VIOS VIOS ent0 fcs0
Storage
Subsystem
Figure 5-1 Live Partition Mobility, resource check
2. If this check is successful, a shell LPAR is created on System #2, as shown in Figure 5-2.

Suspended
IBM i Client
Partition
1 Shell Partition
M M M M M M M Create shell LPAR on M
target system
A LIN1 LIN1
DC01 CMN01 CMN01 DC01
HMC
VLAN VLAN
VASI vhost0 ent1 ent1 VASI

SEA (if) (if) SEA
Storage
Subsystem
Figure 5-2 Live Partition Mobility, shell LPAR created
3. Virtual SCSI or NPIV devices are created in the virtual target VIOS server and in the shell
partition to give the partition access to the storage subsystem that is used by System #1,

Suspended Partition
IBM i Client 1 Shell Partition
M M M M M M M Create virtual SCSI M
devices
A LIN1 LIN1 A

HMC
VLAN VLAN
VASI vhost0 ent1 ent1 vhost0 VASI

Service vtscsi0 vtscsi0 Service
SEA (if) (if) SEA
Storage
Subsystem
Figure 5-3 Live Partition Mobility, virtual SCSI devices created
Chapter 5. Live Partition Mobility 159

4. Main memory is migrated from System #1 to System #2, as shown in Figure 5-4. Users
can still work on System #1 when this process is occurring. Whenever a page from main
memory is changed on System #1 after it was moved to System #2, that page must be
moved from System #1 to System #2 again. Therefore, it is considered a best practice to
start a partition migration during times of low activity on the system.

Suspended
IBM i Client
Partition
1 Shell
IBM i Partition
Client 1
Start migration of
M M M M M
main memory
A LIN1 LIN1 A

HMC
VLAN VLAN

SEA (if) (if) SEA
M M
Storage
Subsystem
Figure 5-4 Live Partition Mobility, memory migration
5. When the system reaches a state where only a small number of memory pages were not
migrated, System #1 is suspended, as shown in Figure 5-5. From this point, no work is
possible on System #1. Users notice that their transactions seem to be frozen. The
remaining memory pages are then moved from System #1 to System #2.

Suspended Partition Shell
IBM i Partition
Client 1
Once enough memory
M M M M M M M
pages have been
moved, suspend the
source system
A LIN1 LIN1 A

HMC
VLAN VLAN

SEA (if) (if) SEA
Storage
Subsystem
Figure 5-5 Live Partition Mobility, suspend of source system

6. After the memory pages migration is finished, the original LPAR definition on System #1 is
deleted with the virtual I/O definitions in the Virtual I/O Server, as shown in Figure 5-6. The
target system resumes and user transactions start to run again.
All IBM i jobs are moved over to the new systems. Users experience a delay in response
time when the system is suspended. However, because jobs are still active on the system
after the migration finishes, uses can continue to work without having to log on to the
system again.
Finish the migration Shell

IBM i Partition
Client 1
and remove the M M M M M M M

original LPAR
definitions
LIN1 A
CMN01 DC01
HMC
VLAN VLAN
VASI ent1 ent1 vhost0 VASI

SEA (if) (if) SEA
fcs0 ent0 VIOS ent0 fcs0

VIOS
Storage
Subsystem
Figure 5-6 Live Partition Mobility, finished status
5.2 License considerations with Live Partition Mobility

When partitions are moved from one system to another, you must ensure that you still adhere
to license requirements in your environment. For IBM i licenses, the following rules apply:
򐂰 When an LPAR is moved to another system permanently, considerations when using Live
Partition Mobility are the same as when a manual move to another system is performed.
IBM i entitlements are licensed to the hardware serial number. Therefore, you need a new
license on the system the LPAR is migrated to or you can transfer entitlements from the
old to the new system under certain circumstances. Licenses for IBM i licensed programs
can be moved to the new system.
򐂰 When an LPAR is moved to another system temporarily, the following scenarios are
possible:
– All necessary licenses are available on both systems. No special considerations need
to be taken.
– The secondary system is a CBU edition. In this case, normal CBU regulations apply.
򐂰 No licenses are available on the secondary system. In this case, the 70-day grace period
applies. You can use the system for 70 consecutive days. When you are performing
multiple Live Partition Mobility operations, this 70-day grace period restarts after each
migration. To work in this way, licenses must be available on the source system. Source
and target systems must belong to the same company or must be leased by the same
company. In addition, the source system must be in a higher or equal processor group as
the target system. To receive support when working on the target system, it is
recommended to have at least one IBM i license with a valid software maintenance
contract on that system.
5.3 Setting up Live Partition Mobility

The use of Live Partition Mobility requires several preparation steps that are described in this
section. In addition, there are several error messages and warnings that might appear during
a Live Partition Migration operation. The most common messages and warnings and their
causes are explained in this section.
5.3.1 Requirements
The following minimum prequisites for the use of Live Partition Mobility in an IBM i
environment must be met on the source system and target system:
򐂰 POWER7® server with firmware level 740_40 or 730_51, or higher
򐂰 HMC V7R7.5 or later
򐂰 VIOS 2.2.1.4 (FP25-SP2) or later
򐂰 IBM i 7.1 Technology Refresh 4 or later
򐂰 PowerVM Enterprise Edition
򐂰 All VIOS virtual adapters assigned as required
򐂰 If NPIV is used, both logical WWPNs of a virtual FC server adapter must be zoned
because the second WWPN is used on destination server.
򐂰 VLAN ID for IBM i client to be migrated must be identical on source and target system.

5.3.2 Preparation
Live Partition Mobility is enabled when PowerVM Enterprise Edition is installed on the server,
as shown in the HMC server properties window in Figure 5-7. You can also see that the
system is enabled for IBM i partition mobility.
Figure 5-7 Live Partition Mobility, IBM i partition mobility capable
All I/O for the LPAR that you want to move to another system must be virtual and all disk
storage for that LPAR must be attached externally by using VSCSI or NPIV. Access to that
storage must be shared between VIO servers on the source and the target system.
No physical adapters can be assigned to the IBM i LPAR. The IBM i LPAR must be set to
Restricted I/O, which is done in the Properties section of the LPAR, as shown in Figure 5-8.
Setting this parameter requires an IPL of the LPAR. It prevents the addition of physical
adapters to the LPAR.
Figure 5-8 Live Partition Mobility, restricted I/O partition

You must specify which VIO Server on the source and on the target LPAR is designated as
the mover service partition (MSP). This enables the specified Virtual I/O Server partition to
asynchronously extract, transport, and install partition state information. Mover service
partitions are not used for inactive migrations. If the MSP flag is not set for any VIO server on
the system, the error message that is shown in Figure 5-9 opens during the Live Partition
Mobility validation.
Figure 5-9 Live Partition Mobility, no mover service partition specified
The attribute for the mover service partition is set in the properties of the VIO server LPAR, as
shown in Figure 5-10. The attribute becomes active immediately and does not require an IPL
of the LPAR.
Figure 5-10 Live Partition Mobility, mover service partition flag
All VIO servers that are taking part in the Live Partition Mobility activity must have a working
RMC connection. If VIO servers are cloned by using copy operations, this is not the case
because their ct_node_id is not unique. In this case, the error message that is shown in
Figure 5-11 is displayed. For more information about for the steps that are required to set the
ct_node_id, see 3.7.2, “Restarting the RMC connection” on page 95.
Figure 5-11 Live Partition Mobility, VIO server without RMC connection
Ensure that the memory region sizes of the source and the target system are identical. If this
setting is different on the systems that are involved, the validation process displays the error
message that is shown in Figure 5-12. If you must change this setting, open the ASMI
interface for the system that you want to change. The value can be found by clicking
Performance Setup  Logical Memory Block Size. Changing the memory region size
requires a complete shutdown of the server.
Figure 5-12 Live Partition Mobility, error for different memory region sizes

5.3.3 Live Partition Mobility migration
A Live Partition Mobility migration consists of the following general steps (the validation and
the actual migration):
1. A Live Partition Mobility validation is started from the HMC GUI, as shown in Figure 5-13.
Figure 5-13 Live Partition Mobility, starting validation
2. Select the destination system to which you want to move the LPAR, as shown in
Figure 5-14. Click Validate.
Figure 5-14 Live Partition Mobility, selecting the destination system
3. After the validation finishes without errors, you can start the migration by clicking Migrate
as shown in Figure 5-15. If there are multiple VIO servers available on the target system to
provide I/O for the migrated LPAR, the Live Partition Mobility process provides a
suggestion for the virtual storage assignment. This predefined setting can also be
changed.
Figure 5-15 Live Partition Mobility, starting migration
4. During the migration, a status window displays the progress, as shown in Figure 5-16.
Figure 5-16 Live Partition Mobility, migration status

5. On IBM i, the messages in the QSYSOPR message queue (as shown in Figure 5-17)
indicate that a partition mobility action is in progress. The license information is displayed
after a successful migration if a license key for the target system is not available.
Display Messages
System: Z1436C23
Queue . . . . . : QSYSOPR Program . . . . : *DSPMSG
Library . . . : QSYS Library . . . :
Severity . . . : 60 Delivery . . . : *HOLD
Type reply (if required), press Enter.

Partition suspend request in progress.
Partition resumed after migration.
License key will not be valid in 70 days on 01/01/14.
License key will not be valid in 70 days on 01/01/14.
Figure 5-17 Live Partition Mobility QSYSOPR messages
5.3.4 Hints and tips for the validation process

During the validation, the following errors and warnings might appear:
򐂰 Virtual optical devices must be deallocated before a Live Partition Mobility operation can
start. Reset the virtual IOP #290A from IBM i SST and unconfigure the virtual optical
target device in VIOS by using the following command:
rmdev -dev vtopt0 -ucfg
Otherwise, the error message that is shown in Figure 5-18 is displayed.
Figure 5-18 Live Partition Mobility error message if the virtual optical device is not deallocated
򐂰 Tape devices must be varied off from the IBM i side. If a tape is still varied on, this does not
result in an error message during the validation phase. During the actual migration, the
error message that is shown in Figure 5-19 is displayed and the Live Partition Mobility
operation stops.
Figure 5-19 Live Partition Mobility error message if tape device is varied on
In addition, the information that is shown in Figure 5-20 can be found in the QSYSOPR
message queue of the LPAR that you attempted to move.
Additional Message Information
Message ID . . . . . . : CPF67AC Severity . . . . . . . : 40

Message type . . . . . : Information
Date sent . . . . . . : 10/23/13 Time sent . . . . . . :
11:43:05
Message . . . . : Function *MIGRATION not supported.

Cause . . . . . : The parameter, value or command *MIGRATION was
requested but is not supported for device TAP01.
Recovery . . . : Do not specify the command, parameter or value that is
not supported.
Figure 5-20 Live Partition Mobility QSYSOPR message if tape device not varied off

򐂰 A warning message is shown during validation if the slot numbers for server virtual Fibre
Channel adapters or server virtual SCSI adapters cannot be retained, as shown in
Figure 5-21. This is a warning only and it does not stop the migration process.
Figure 5-21 Live Partition Mobility warning message regarding slot numbers
򐂰 If you mapped two virtual Fibre Channel adapters to one physical port for the LPAR on the
source system, the error message that is shown in Figure 5-22 is displayed. The same
error message appears if the Live Partition Mobility operation results in a setup on the
target system where two virtual Fibre Channel adapters are mapped to one physical port.
Figure 5-22 Live Partition Mobility error message for physical device location not found
6
Chapter 6. BRMS and FlashCopy

In times where business demands for ever rising availability levels on systems with a high
workload, backup windows are closing fast. With Backup, Recovery and Media Services
(BRMS) and FlashCopy, an independent ASP (IASP) can be saved without a diminishing
effect on applications and disk performance. With PowerHA, FlashCopy management is done
from the IBM i partition for ease of use.
This chapter provides information about how an IASP is suspended for a brief period, the
FlashCopy is taken, and the BRMS setup is done.

򐂰 Introduction to PowerHA, BRMS, and FlashCopy scenario
򐂰 PowerHA and FlashCopy
򐂰 Backup Recovery, and Media Services (BRMS)

6.1 Introduction to PowerHA, BRMS, and FlashCopy scenario
A common setup for this environment is a three-node cluster, as shown in Figure 6-1. High
availability is provided between node A and node B by using Metro Mirror on the storage
system. In addition, a FlashCopy point in time copy is taken from the LUNs that make up the
IASP. This copy of the IASP is attached to the third node in the cluster. Data in that copy of the
IASP is saved to tape. By using BRMS, these saves can then be restored to cluster nodes A
and B.
Node C Node A Node B

BRMS Network
Power HA
Backup LPAR z1436b22 ASP1 Primary LPAR z1436b23 ASP1 High Availability LPAR z1436c23 ASP1
Flash Copy Target Production MetroMirror

Disk 1 Disk 3 Disk 1 Disk 3 Disk 1 Disk 3
iASP1 FlashCopy iASP1 MetroMirror iASP1
Disk 2 Disk 4 Disk 2 Disk 4 Disk 2 Disk 4
Re
Bac sto tore
kup r Res
e
Figure 6-1 PowerHA, FlashCopy, and BRMS scenario
6.2 PowerHA and FlashCopy

The use of FlashCopy in an IBM i environment requires a tight integration between IBM i and
the storage system that is running the FlashCopy operation. There are steps that must be
taken on the IBM i side, such as quiescing the application to reach a consistent state of data
on the disks. When this is accomplished, the FlashCopy is taken on the storage system. After
this operation is finished, access to applications and data on the IBM i side must be started
again.
PowerHA provides this integration. There are IBM i commands available with PowerHA that
communicate with the storage system by using Secure Shell (SSH) communication to run the
required operations.
6.2.1 Preparation for FlashCopy on the Storwize side

Preparation on the Storwize family side is easy because most of the configuration tasks are
taken over by IBM PowerHA SystemMirror for i. You must create only the required LUNs and
attach them to the host you want to use them on.
Preparation must be done on another LPAR. You cannot attach this copy of the IASP to your
backup node. The FlashCopy consistency groups and FlashCopy mappings are automatically
created when you start a FlashCopy session from the IBM i.
6.2.2 Changes in PowerHA to add the third node into the cluster
As shown in the example in Figure 6-1 on page 174, the two-node cluster for high availability
is already set up with node z1436b23 as the primary system and node z1436c23 as the
secondary system. To use FlashCopy and BRMS, complete the following steps to extend that
configuration to include a third node z1436b22:
1. On the third node, create a device description for the IASP, as shown in the following
example:
2. On one of the existing cluster nodes, add the third node into the cluster, as shown in the
following example:
ADDCLUNODE CLUSTER(SVC_CLU) NODE(Z1436B22 ('9.5.216.230' '192.168.10.22'))
3. Add the new cluster node into the existing device domain, as shown in the following
example:
4. Add an SAN Volume Controller copy description that points to the newly created LUNs, as
shown in the following example:
ADDSVCCPYD ASPCPY(FLASHSVCB) ASPDEV(IASP1) CRG(*NONE) SITE(*NONE)
NODE(Z1436B22) SVCHOST(powerha '/QIBM/UserData/HASM/hads/.ssh/id_rsa'
'9.5.216.236') VRTDSKRNG((8 11 *ALL))
Note: For the target copy description that is to be used for FlashCopy, the cluster
resource group and cluster resource group site must be *NONE. The node identifier
must be set to the cluster node name that owned the target copy of the IASP.
A FlashCopy can now be taken by using the STRSVCSSN command, as shown in Figure 6-2.
Start SVC Session (STRSVCSSN)
Type choices, press Enter.
Session . . . . . . . . . . . . > FLASH1 Name

Session type . . . . . . . . . . > *FLASHCOPY *METROMIR, *GLOBALMIR...
Device domain . . . . . . . . . * Name, *
ASP copy:
Preferred source . . . . . . . IASP_MM Name
Preferred target . . . . . . . FLASHSVCB Name
+ for more values
Incremental flash . . . . . . . *NO *NO, *YES
Copy rate . . . . . . . . . . . 0 0-100
Cleaning rate . . . . . . . . . 0 0-100
Grain size . . . . . . . . . . . 256 256, 64
Consistency group . . . . . . . *GEN
Reverse consistency group . . . *GEN
Bottom
F3=Exit F4=Prompt F5=Refresh F12=Cancel F13=How to use this display
F24=More keys
Figure 6-2 Using the STRSVCSSN command to start FlashCopy
Chapter 6. BRMS and FlashCopy 175

By using the STRSVCSSN command, the user can create an incremental FlashCopy relationship
and specify a FlashCopy background copy rate and grain size. PowerHA requires the
FlashCopy relationships to be included in a consistency group that is, by default, newly
created by PowerHA on the SVC/V7000 when a FlashCopy session is started. Alternatively,
the user can specify the name for an existing FlashCopy consistency group.
Issuing an End SVC ASP Session (ENDSVCSSN) command, as shown in Figure 6-3, ends the
FlashCopy relationship. The consistency group is deleted by default.
End SVC ASP Session (ENDSVCSSN)
Session . . . . . . . . . . . . FLASH1 Name

Device domain . . . . . . . . . * Name, *
Delete consistency group . . . . *YES *YES, *NO
Bottom
F3=Exit F4=Prompt F5=Refresh F12=Cancel F13=How to use this display
F24=More keys
Figure 6-3 Using the ENDSVCSSN command to end FlashCopy
6.3 Backup Recovery, and Media Services (BRMS)

Backup, Recovery, and Media Services (BRMS) for i is the IBM strategic solution for planning
and managing the backup of your IBM i server.
BRMS provides the IBM i server with support for policy-oriented setup and the running of
backup, recovery, archive, and other removable-media-related operations. BRMS uses a
consistent set of intuitive concepts and operations that can be used to develop and implement
a backup strategy that is tailored to your business requirements.
For more information about BRMS, see Backup Recovery and Media Services for OS/400: A
Practical Approach, SG24-4840.
6.3.1 Setting up the BRMS network group

To share the BRMS database information about the media classes, available storage
locations, and media policies, all PowerHA nodes must be brought into a BRMS network
configuration. Complete the following steps:
1. Log on to the IBM i LPAR and open the BRMS policy menu by using the following
command:
WRKPCYBRM *SYS
2. From the System Policy menu, select option 4, Change network group.
3. In the Change Network Group display (see Figure 6-4), change the communication
method to *IP and add your IBM i LPARs as nodes to the BRMS network. The host name
must be reachable in the network. Add a host table entry, if necessary.
Tip: If the add operation fails, change the DDM attributes to allow a connection without
password by using the following command:
CHDDMTCPA PWDRQD(*USRID)
Change Network Group Z1436B22 APPN
Network group . . . . : *MEDINV Position to . . . . :

FlashCopy state . . . : *ENDPRC
Communication method : *IP *AVAIL, *IP, *SNA, *NONE
Notify period . . . . : 30 30-99999 seconds
Type options, press Enter.
1=Add 4=Remove 8=Set time
Remote Local Remote System Network

Opt Location Name Network ID Status Status
1 Z1436B23 APPN
Z1436C23 APPN Online Active
Bottom
F3=Exit F5=Refresh F11=BRMS Media Information F12=Cancel

Figure 6-4 Adding node z1436b23 to the BRMS network
4. Import the BRMS database information from the leading BRMS system to the secondary
partitions by using the following command (ignore all messages):
INZBRM OPTION(*NETSYS) FROMSYS(APPN.Z1436B22)
5. From the system that backs up the IASP, define which systems receive the save
information and define that they also own the save history. If the option to change the save
history owner is implemented, it appears as though the target system performed the
backup. This is useful for the PRTRCYRPT and STRRCYBRM commands. Complete the following
steps:
a. To add a specific system sync, run the following command to change the system name
to make it look as though the backup was done by this system and synchronize the
reference date and time:
CALL QBRM/Q1AOLD PARM('HSTUPDSYNC' '*ADD' 'Z1436B23' 'APPN' 'IASP1'
'*CHGSYSNAM')
b. To add a specific system sync, run the following command to keep the name of who did
the backup and synchronize the reference date and time:
CALL QBRM/Q1AOLD PARM('HSTUPDSYNC' '*ADD' 'Z1436C23' 'APPN' 'IASP1'
'*NORMAL')
c. To display what is setup, run the following command:
CALL QBRM/Q1AOLD PARM('HSTUPDSYNC' '*DISPLAY')
d. To remove a specific system, use the following command:
CALL QBRM/Q1AOLD PARM('HSTUPDSYNC' '*REMOVE' 'Z1436C23' 'APPN' 'IASP1')

6.3.2 Configuring BRMS media classes and media policies
BRMS divides tape media into different classes. These classes must be added first because
they are used later in the media policy. When the media policy is created, groups of tapes are
divided by these classes. If there is a fixed set of tapes for daily, weekly, and monthly backups,
create multiple policies.
Complete the following steps to create a media class and a media policy:
1. Run the Work with Media Classes command WRKCLSBRM to open the media classes menu.
2. Add the media class, as shown in Figure 6-5. Repeat this step as necessary for your
normal BRMS save routines.
Add Media Class
Media class . . . . . . . . . . . . DAILYIASP Name

Density . . . . . . . . . . . . . . *ULTRIUM5 F4 for list
Media capacity . . . . . . . . . . *DENSITY *DENSITY, Number nnnnn.nn
Unit of measure . . . . . . . . . 1=KB, 2=MB, 3=GB
Mark for label print . . . . . . . *NONE *NONE, *MOVE, *WRITE
Label size . . . . . . . . . . . . 1 1=6 LPI, 2=8 LPI, 3=9 LPI
Label output queue . . . . . . . . *SYSPCY Name, *SYSPCY, *PRTF
Library . . . . . . . . . . . . . Name, *LIBL
Shared media . . . . . . . . . . . *YES *YES, *NO
Write once media . . . . . . . . . *NO *YES, *NO
Text . . . . . . . . . . . . . . . used for save of IASP1
Figure 6-5 Adding a media class for IASP save
3. Run the WRKPCYBRM TYPE(*MED) command to add a new policy for the save operation, as
shown in Figure 6-6. This example uses a retention period of one day. In most cases, you
might prefer longer retention periods.
Create Media Policy
Media policy . . . . . . . . . . . DAILYIASP Name

Retention type . . . . . . . . . . 2 1=Date, 2=Days,
3=Versions, 4=Permanent
Retain media . . . . . . . . . . 1 Date, Number
Deleted library retention. . . . *NONE Number, *NONE
Move policy . . . . . . . . . . . *NONE Name, *NONE, *ADSM, F4
Media class . . . . . . . . . . . DAILYIASP Name, *SYSPCY, *ADSM, F4
Storage location . . . . . . . . . *ANY Name, *ANY, F4 for list
Save to save file . . . . . . . . *NO *YES, *NO
ASP for save files . . . . . . . *SYSTEM Name, *SYSTEM, 1-32
Save file retention type . . . . 4 1=Date, 2=Days,
3=Permanent, 4=None
Retain save files . . . . . . *NONE Date, Number, *NONE
ASP storage limit . . . . . . . *SYS *SYS, 1-99
Secure media . . . . . . . . . . . *NO *YES, *NO, *ADSM
More...
Required volumes . . . . . . . . . *NONE *NONE, 1-9999

Mark volumes for duplication . . . *NO *NO, *YES
Mark history for duplication . . . *NO *NO, *YES
Text . . . . . . . . . . . . . . . used for save of IASP1
Figure 6-6 Add media policy for IASP save
6.3.3 Configuring the BRMS control group

Control groups define logical groups of libraries and objects that possess similar backup,
retention, and recovery requirements. In addition to allowing you to define the order in which
backup, archive, and recovery processing occurs, control groups provide for special related
actions, such as tape loads, processing subsystems, and job queues. Control groups provide
exits for user-defined processing during the backup cycle.

Figure 6-7 shows in which order the BRMS control groups that are created in this section
must be run. This can be done with the job scheduler or by using the Run Remote Command
(RUNRMTCMD) command.
Figure 6-7 Backup run for FlashCopy IASP
Use the Work with Backup Control Groups command (WRKCTLGBRM) to create BRMS control
groups. Complete the following steps:
1. As shown in Example 6-1, create control groups on the primary and high availability node,
which is used to create and remove the FlashCopy of the IASP1. Also, the recovery
reports are printed on the primary node.
Example 6-1 Creating two BRMS control groups with CRTIASPCPY and RMVIASPCPY
CRTIASPCPY *EXIT CHGASPACT ASPDEV(IASP1) OPTION(*SUSPEND) SSPTIMO(3)
*EXIT STRSVCSSN SSN(FLASH1) TYPE(*FLASHCOPY) ASPCPY((IASP_MM FLASHSVCB))
*EXIT CHGASPACT ASPDEV(IASP1) OPTION(*RESUME)
RMVIASPCPY *EXIT ENDSVCSSN SSN(FLASH1)

*EXIT SETASPGRP ASPGRP(IASP1)
*EXIT STRMNTBRM RUNCLNUP(*NO) RTVVOLSTAT(*NO) PRTEXPMED(*NO) PRTBKUACT(*NO) PRTRCYRPT(*RCYANL)
2. On the backup node, create the control group that is used to back up the IASP copy, as
shown in Example 6-2.
Example 6-2 Backup node control group

DAILYIASP *EXIT DAILYIASP DAILYIASP VRYCFG CFGOBJ(IASP1) CFGTYPE(*DEV) STATUS(*ON)
*ALLUSR IASP1 DAILYIASP DAILYIASP *YES
*LINK IASP1 DAILYIASP DAILYIASP *YES
*EXIT DAILYIASP DAILYIASP VRYCFG CFGOBJ(IASP1) CFGTYPE(*DEV) STATUS(*OFF)
*EXIT DAILYIASP DAILYIASP SNDMSG MSG('BACKUPS ARE COMPLETE') TOUSR(*SYSOPR)
3. On the backup LPAR, the production LPAR was added to the list of owners for the saved
object, as described in 6.3.1, “Setting up the BRMS network group” on page 176.
4. Run the BRMS control groups by using Start Backup by using the BRM (STRBKUBRM)
command in the correct order to save the IASP objects.
Figure 6-8 shows that after running the following Start Recovery by using BRM (STRRCYBRM)
command with *LCL as the originating system, all information about the saved IASP objects
from the backup LPAR are available:
STRRCYBRM OPTION(*ALLUSR) ACTION(*RESTORE) FROMSYS(*LCL)
Select Recovery Items Z1436B23
Select action . . . . . : *ALL

Select volume . . . . . :
Type options, press Enter.
1=Select 4=Remove 5=Display 7=Specify object
Saved Save Save Save Parallel Volume File Expire

Opt Item Date Time Type Devices Serial Sequence Date
QRCL00033 10/16/13 15:29:15 *FULL 1FA390 3 10/17/13
QSYS200033 10/16/13 15:29:15 *FULL 1FA390 2 10/17/13
SYSIB00033 10/16/13 15:29:15 *FULL 1FA390 4 10/17/13
TESTJORDAN 10/16/13 15:29:15 *FULL 1FA390 5 10/17/13
Bottom
F3=Exit F5=Refresh F9=Recovery defaults F11=Object View
F12=Cancel F14=Submit to batch F16=Select
Figure 6-8 STRRCYBRM command on the production system

A
Appendix A. PowerHA Tools IBM i

This appendix describes the PowerHA Tools for IBM i offerings and services that are available
from IBM Systems Lab Services.
The PowerHA Tools for IBM i complement and extend the PowerHA and IBM storage
capabilities for high availability (HA) and disaster recover (DR).
The PowerHA Tools for IBM i provide the following benefits:

򐂰 Helps reduce business risk and improve resiliency for critical applications.
򐂰 Simplifies setup and automation of HA, DR, and backup solutions.
򐂰 Reduces the cost of maintaining and regular testing of an HA/DR environment.
򐂰 Facilitates flexible deployment options for single or multi-site protection.
򐂰 Assures consistent deployment by using best practices and experienced consultants.
For more information about PowerHA Tools for IBM i, see the following IBM Systems Lab
Services and Training website:
http://www.ibm.com/systems/services/labservices

PowerHA Tools for IBM i
Table A-1 lists the PowerHA Tools for IBM i that are available from IBM Systems Lab
Services.
Table A-1 PowerHA Tools for IBM i

PowerHA Tools for Capability Benefit
Storwize
DS8000
Internal
IBM i
Smart Assist for Provides operator commands and Simplifies deployment and ongoing X X X
PowerHA on IBM i scripts to supplement PowerHA management of HA for critical IBM i
installation and ongoing operations for applications.
iASP enabled applications.
IASP Copy Services Manager

(Automated recovery with faster IASP-level vary on, no system IPL)
Flashcopy Automates Flashcopy of IASP for daily Increases application availability by X X

offline backup with seamless BRMS reducing or eliminating backup
integration. window for routine daily backups.
LUN-level Switching Simplifies deployment and automates Enables a business continuity X1 2
switching an IASP between IBM i manager to provide a simple, single

cluster nodes in one data center. site HA solution.
Metro Mirror or Simplifies initial deployment and Enables a business continuity X

Global Mirror automates ongoing server and storage manager to provide seamless
management of two-site Metro Mirror or operation of integrated server and
Global Mirror HA or DR solutions. storage operations for two-site HA
Requires IASP enabled applications. and DR.
Metro Global Mirror Extends PowerHA functionality to Enables a business continuity X

(MGM) provide three-site server or storage manager to further lower business
replication solution containing Metro risk and maximize business
Mirror for HA with Global Mirror for DR. resilience for critical applications
Requires IASP enabled applications that require three-site HA and DR
and IBM Tivoli Productivity Center for protection.
Replication.
Full System Copy Services Manager

(Automated recovery, requires full system IPL)
Flashcopy Automates full system Flashcopy for Increases application availability by X X X

daily offline backup with integrated reducing or eliminating backup
support for BRMS without IASP window for online daily backups.
enabled applications. Enables an entry solution while
planning IASP enablement.
Metro Mirror or Simplifies initial deployment and Enables a business continuity X

Global Mirror automates ongoing server and storage manager to provide seamless
management of two-site Metro Mirror or operation of integrated server and
Global Mirror HA or DR solitons without storage operations for HA and DR.
IASP enabled applications. Enables an entry solution while an
IASP enablement is planned.
1 DS8000 support available with PowerHA Tools for IBM i 6.1 or earlier; included in PowerHA SystemMirror 7.1
2
V7000 support included with PowerHA 7.1 TR6
IBM Lab Services Offerings for PowerHA for IBM i
Table A-2 lists the PowerHA for IBM i service offerings that are available from IBM Systems
Lab Services.
Table A-2 PowerHA for IBM i service offerings

PowerHA for IBM i Description
Service Offering
IBM i High Availability Architecture An experienced IBM i consultant conducts a planning and design workshop to
and Design Workshop review solutions and alternatives to meet HA and DR and backup and recovery
requirements. The consultant provides an architecture and implementation plan to
meet these requirements.
PowerHA for IBM i An experienced IBM i consultant reviews network bandwidth requirements for
Bandwidth Analysis implementing storage data replication. IBM reviews I/O data patterns and provides
a bandwidth estimate to build into the business and project plan for clients who are
deploying PowerHA for IBM i.
IBM i Independent Auxiliary An experienced IBM i consultant provides jumpstart services for migrating
Storage Pool (IASP) Workshop applications into an IASP. Training includes enabling applications for IASPs,
clustering techniques, and managing PowerHA and HA and DR solution options
with IASPs.
PowerHA for IBM i An experienced IBM consultant provides services to implement an HA and DR
Implementation Services solution for IBM Power Systems servers with IBM Storage. Depending on specific
business requirements, the end-to-end solution implementation can include a
combination of PowerHA for IBM i and PowerHA Tools for IBM i, and appropriate
storage software, such as Metro Mirror, Global Mirror, or Flashcopy.
Appendix A. PowerHA Tools IBM i 185

Related publications
The publications that are listed in this section are considered particularly suitable for a more
detailed discussion of the topics that are covered in this book.
IBM Redbooks
The following IBM Redbooks publications provide more information about the topics in this
document. Some of the publications that are referenced in this list might be available in
softcopy only:
򐂰 PowerHA SystemMirror for IBM i Cookbook, SG24-7994
򐂰 IBM Power Systems HMC Implementation and Usage Guide, SG24-7491
򐂰 Simple Configuration Example for Storwize V7000 FlashCopy and PowerHA SystemMirror
for i, REDP-4923
򐂰 IBM i and Midrange External Storage, SG24-7668
򐂰 IBM System Storage SAN Volume Controller and Storwize V7000 Replication Family
Services, SG24-7574
򐂰 Backup Recovery and Media Services for OS/400: A Practical Approach, SG24-4840
You can search for, view, download, or order these documents and other Redbooks,
Redpapers, Web Docs, draft, and other materials, at this website:
http://www.ibm.com/redbooks
Other publications
Another relevant source of more information is IBM Real-time Compression Evaluation User
Guide, S7003988, which is available at this website:
http://www-01.ibm.com/support/docview.wss?uid=ssg1S7003988&myns=s028&mynp=OCST3FR7
&mync=E
Online resources
These websites are also relevant as further information sources:
򐂰 IBM System Storage Interoperation Center (SSIC):
http://www-03.ibm.com/systems/support/storage/ssic/interoperability.wssDescript
ion1
򐂰 IBM System Storage SAN Volume Controller:
http://www-03.ibm.com/systems/hk/storage/disk/storwize_v3500/index.html

򐂰 IBM Storwize V7000 and Storwize V7000 Unified Disk Systems:
򐂰 IBM Comprestimator Utility:
http://www-304.ibm.com/support/customercare/sas/f/comprestimator/home.html
򐂰 IBM Systems Workload Estimator:
http://www-912.ibm.com/wle/EstimatorServlet
򐂰 Subsystem Device Driver Path Control Module (SDDPCM) driver:
򐂰 Virtual I/O Server Advisor tool:
https://www.ibm.com/developerworks/community/wikis/home?lang=en#!/wiki/Power%20
Systems/page/VIOS%20Advisor
Help from IBM

IBM Support and downloads:
http://www.ibm.com/support
IBM Global Services:

http://www.ibm.com/services
IBM i and IBM Storwize Family: A
Practical Guide to Usage Scenarios
IBM i and IBM Storwize Family: A Practical Guide to Usage Scenarios
IBM i and IBM Storwize Family: A Practical Guide to Usage Scenarios
(0.2”spine)
0.17”<->0.473”
90<->249 pages
Back cover ®
IBM i and IBM Storwize Family

A Practical Guide to Usage Scenarios
®
VIOS explained from The use of external storage and the benefits of virtualization became a
topic of discussion in the IBM i area during the last several years. The INTERNATIONAL
an IBM i perspective
question tends to be, what are the advantages of the use of external TECHNICAL
storage that is attached to an IBM i environment as opposed to the use SUPPORT
Using PowerHA of internal storage. The use of IBM PowerVM virtualization technology ORGANIZATION
together with IBM to virtualize Power server processors and memory also became
Storwize Family common in IBM i environments. However, virtualized access to external
storage and network resources by using a VIO server is still not widely
Configuring BRMS in used.
a high available This IBM Redbooks publication gives a broad overview of the IBM BUILDING TECHNICAL
solution Storwize family products and their features and functions. It describes INFORMATION BASED ON
the setup that is required on the storage side and describes and PRACTICAL EXPERIENCE
positions the different options for attaching IBM Storwize family
products to an IBM i environment. Basic setup and configuration of a IBM Redbooks are developed
VIO server specifically for the needs of an IBM i environment is also by the IBM International
described.
Technical Support
Organization. Experts from
In addition, different configuration options for a combined setup of IBM IBM, Customers and Partners
PowerHA SystemMirror for i and the Storwize family products are from around the world create
described and positioned against each other. Detailed examples are timely technical information
provided for the setup process that is required for these environments. based on realistic scenarios.
Specific recommendations
The information that is provided in this book is useful for clients, IBM are provided to help you
Business Partners, and IBM service professionals who need to implement IT solutions more
understand how to install and configure their IBM i environment with effectively in your
attachment to the Storwize family products. environment.
For more information:

ibm.com/redbooks
SG24-8197-00 ISBN 0738439223

Storwize and IBM I PDF

Uploaded by

Copyright:

Available Formats

Storwize and IBM I PDF

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

Storwize and IBM I PDF

Uploaded by

Copyright:

Available Formats

What are the different attachment options for connecting IBM Storwize family storage to an IBM i environment?

What are the different attachment options for connecting IBM Storwize family storage to an IBM i environment?

What are the basics of the IBM Storwize family?

What are the basics of the IBM Storwize family?

Front cover

IBM i and IBM Storwize Family

VIOS explained from an IBM i

Using PowerHA together with IBM

Configuring BRMS in a high

IBM i and IBM Storwize Family: A Practical

First Edition (April 2014)

This edition applies to IBM i 7.1 Technology Refresh (TR) 7.

© Copyright International Business Machines Corporation 2014. All rights reserved.

Chapter 1. Introduction to IBM Storwize Family for IBM i . . . . . . . . . . . . . . . . . . . . . . . . 5

Chapter 2. Configuring IBM Storwize V7000 with IBM i. . . . . . . . . . . . . . . . . . . . . . . . . 17

Chapter 3. Setting up VIOS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 63

© Copyright IBM Corp. 2014. All rights reserved. iii

Chapter 4. PowerHA and the IBM Storwize Family . . . . . . . . . . . . . . . . . . . . . . . . . . . . 97

Chapter 5. Live Partition Mobility . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 157

Chapter 6. BRMS and FlashCopy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 173

Appendix A. PowerHA Tools IBM i. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 183

iv IBM i and IBM Storwize Family: A Practical Guide to Usage Scenarios

Related publications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 187

© Copyright IBM Corp. 2014. All rights reserved. vii

The following terms are trademarks of other companies:

In addition, different configuration options for a combined setup of IBM PowerHA®

Sabine Jordan is a Consulting IT Specialist in IBM Germany.

© Copyright IBM Corp. 2014. All rights reserved. 1

Rodrigo Jungi Suzuki is SAN Storage specialist at IBM Brazil

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

2 IBM i and IBM Storwize Family: A Practical Guide to Usage Scenarios

Stay connected to IBM Redbooks

Chapter 1. Introduction to IBM Storwize

The following topics are covered in this chapter:

© Copyright IBM Corp. 2014. All rights reserved. 5

It is reasonable to use internal storage for the following reasons:

However, internal disks can be annoying for the following reasons:

6 IBM i and IBM Storwize Family: A Practical Guide to Usage Scenarios

1.2.1 Direct attached

Chapter 1. Introduction to IBM Storwize Family for IBM i 7

1.2.3 vSCSI attached

8 IBM i and IBM Storwize Family: A Practical Guide to Usage Scenarios

1.2.4 N-Port ID Virtualization (NPIV) attached

Chapter 1. Introduction to IBM Storwize Family for IBM i 9

1.3 Basics of IBM Storwize family

10 IBM i and IBM Storwize Family: A Practical Guide to Usage Scenarios

Figure 1-5 SAN Volume Controller front view

Figure 1-6 shows the front view of Storwize V3700.

Figure 1-6 Storwize V3700 front view

Chapter 1. Introduction to IBM Storwize Family for IBM i 11

Figure 1-7 Storwize V7000 front view

12 IBM i and IBM Storwize Family: A Practical Guide to Usage Scenarios

Figure 1-8 IBM SAN Volume Controller Overview panel

Table 1-1 IBM Storwize videos

Storwize Family http://www.youtube.com/watch?v=M0hIfLKUk1U

Storwize V3700 http://www.youtube.com/watch?v=IdxnH9YWF9s

Storwize V7000 http://www.youtube.com/watch?v=S2p4IQxIiec&feature=c4-

Chapter 1. Introduction to IBM Storwize Family for IBM i 13

IBM Real-time Compression and http://www.youtube.com/watch?v=rgKj75kn2J0