IBM SmartCloud Virtual Storage Center

IBM SmartCloud Virtual Storage Center
Front cover
IBM SmartCloud
Virtual Storage Center
Explore virtualization, storage, and
cloud for your infrastructure
Accelerate business insights with
advanced analytics
Follow along with sample
guided use cases
Karen Orlando
Joseph Gatti
Joao Marcos Leite
Thirumalainambi Murugesh
Danijel Paulin
Socheat Sou
ibm.com/redbooks
International Technical Support Organization
IBM SmartCloud Virtual Storage Center
February 2015
SG24-8239-00
Note: Before using this information and the product it supports, read the information in “Notices” on
page vii.
First Edition (February 2015)
This edition applies to IBM SmartCloud Virtual Storage Center Version 5.2.
© Copyright International Business Machines Corporation 2015. 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
IBM Redbooks promotions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ix
Preface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . iii
Authors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . iv
Now you can become a published author, too! . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . vi
Comments welcome. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . vii
Stay connected to IBM Redbooks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . vii
Chapter 1. Introductory concepts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
1.1 Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
1.2 Software-defined environment (SDE) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
1.2.1 IT SDE infrastructure components . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
1.2.2 Software-defined storage overview. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
1.2.3 IBM SmartCloud Virtual Storage Center (VSC) . . . . . . . . . . . . . . . . . . . . . . . . . .
1.3 The changing landscape of IT environments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
1.4 Cloud computing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
1.4.1 Storage cloud . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
1.4.2 Benefits of storage cloud . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
1.4.3 IBM SmartCloud Virtual Storage Center and storage cloud . . . . . . . . . . . . . . . . .
13
14
15
16
17
19
20
21
22
23
23
Chapter 2. Virtual Storage Center (VSC). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2.1 IBM SmartCloud Virtual Storage Center overview . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2.2 Storage hypervisor . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2.3 IBM SmartCloud Virtual Storage Center component model . . . . . . . . . . . . . . . . . . . . .
2.3.1 Storage management . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2.3.2 Storage virtualization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2.3.3 Application-aware data protection. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2.4 IBM SmartCloud Virtual Storage Center features . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2.4.1 Efficient by design . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2.4.2 Self-optimizing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2.4.3 Cloud agility . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2.5 IBM SmartCloud Virtual Storage Center interfaces. . . . . . . . . . . . . . . . . . . . . . . . . . . .
2.5.1 OpenStack software . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2.5.2 IBM SmartCloud Storage Access . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2.5.3 IBM Cloud Orchestrator . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2.5.4 VMware . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2.6 IBM Virtual Storage Center offerings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2.6.1 License model overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2.6.2 VSC License: Enterprise deployment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2.6.3 VSC Entry License . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2.6.4 VSC for Storwize Family license . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
27
28
28
30
30
33
36
37
37
38
39
40
41
42
45
46
51
52
54
55
55
Chapter 3. Storage management . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
3.1 Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
3.2 Scenarios and installation considerations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
3.2.1 Installing new Tivoli Storage Productivity Center . . . . . . . . . . . . . . . . . . . . . . . . .
57
58
63
64
© Copyright IBM Corp. 2015. All rights reserved.
iii
3.2.2 Upgrading existing Tivoli Storage Productivity Center installation to
Tivoli Storage Productivity Center VSC edition . . . . . . . . . . . . . . . . . . . . . . . . . .
3.2.3 Use case: Replication management, installing Tivoli Storage
Productivity Center in DR environments. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
3.3 Advanced storage analytics overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
3.3.1 Cloud configuration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
3.3.2 Storage provisioning . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
3.3.3 Optimization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
3.4 SAN and storage management . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
3.5 Performance management . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
3.6 Advanced monitoring and alerting. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
67
76
76
77
81
82
84
86
87
Chapter 4. Storage virtualization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 89
4.1 External storage virtualization and software-defined storage . . . . . . . . . . . . . . . . . . . . 90
4.1.1 Product overview and integration with IBM SmartCloud Virtual Storage Center . 90
4.2 Benefits of storage virtualization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 91
4.2.1 Operational efficiency . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 91
4.2.2 Improved return of investment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 91
4.2.3 Enhanced data availability . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 92
4.2.4 Transparent data mobility . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 93
4.3 Storwize software stack . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 93
4.4 SAN Volume Controller component overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 94
4.5 Storwize family component overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 98
4.6 When to choose SAN Volume Controller or Storwize family. . . . . . . . . . . . . . . . . . . . 100
iv
Chapter 5. Application-aware data protection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
5.1 FlashCopy Manager overview. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
5.2 Data protection and recovery capabilities . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
5.2.1 FlashCopy Manager: Supported platforms . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
5.2.2 Components . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
5.2.3 FlashCopy Manager Windows overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
5.2.4 FlashCopy Manager for VMware . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
5.2.5 FlashCopy Manager for UNIX and Linux . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
5.2.6 IBM Tivoli Storage Manager integration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
5.3 Use case: FlashCopy Manager for VMware custom application support . . . . . . . . . .
103
104
104
105
105
106
106
107
107
108
Chapter 6. Provisioning . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
6.1 Provisioning overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
6.1.1 Automated provisioning . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
6.1.2 Cloud provisioning . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
6.1.3 Tivoli Storage Productivity Center: Benefits of using agentless server. . . . . . . .
6.1.4 Planning for provisioning. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
6.1.5 Organizing your storage environment into tiers . . . . . . . . . . . . . . . . . . . . . . . . .
6.1.6 Service classes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
6.1.7 Capacity pools . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
6.2 Use case: Provisioning by using Tivoli Storage Productivity Center vSphere plug-in.
141
142
142
142
144
147
147
149
161
163
Chapter 7. Storage optimization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
7.1 Storage optimization functions and processes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
7.2 Balance Pools . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
7.3 Analyze Tiering . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
7.4 Volume transformation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
7.5 Use case: Balance Pools . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
7.6 Use case: Analyze Tiering . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
171
172
178
185
197
200
210
IBM SmartCloud Virtual Storage Center
7.7 Use case: Volume transformation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 219
Appendix A. Tivoli Storage FlashCopy Manager and DB2 scripts . . . . . . . . . . . . . . . 227
A.1 Scripts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 228
Related publications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
IBM Redbooks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Other publications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Online resources . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Help from IBM . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
237
237
237
237
238
Contents
v
vi
IBM SmartCloud Virtual Storage Center
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. 2015. 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:
Active Cloud Engine®
AIX®
Bluemix™
Cognos®
DB2®
DS4000®
DS5000™
DS8000®
Easy Tier®
FlashCopy®
FlashSystem™
Global Technology Services®
GPFS™
HyperSwap®
IBM®
IBM FlashSystem®
IBM SmartCloud®
Jazz™
Passport Advantage®
ProtecTIER®
pureScale®
Real-time Compression™
Redbooks®
Redbooks (logo)
®
Storwize®
System Storage®
Tivoli®
Tivoli Enterprise Console®
WebSphere®
XIV®
The following terms are trademarks of other companies:
SoftLayer, and SoftLayer device are trademarks or registered trademarks of SoftLayer, Inc., an IBM Company.
Intel, Intel Xeon, Intel logo, Intel Inside logo, and Intel Centrino logo are trademarks or registered trademarks
of Intel Corporation or its subsidiaries in the United States and other countries.
Linux is a trademark of Linus Torvalds in the United States, other countries, or both.
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 SmartCloud Virtual Storage Center
IBM REDBOOKS PROMOTIONS
IBM Redbooks promotions
Find and read thousands of
IBM Redbooks publications
Search, bookmark, save and organize favorites
Get up-to-the-minute Redbooks news and announcements
Link to the latest Redbooks blogs and videos
Download
Now
Android
iOS
Get the latest version of the Redbooks Mobile App
Promote your business
in an IBM Redbooks
publication
®
Place a Sponsorship Promotion in an IBM
Redbooks publication, featuring your business
or solution with a link to your web site.
®
Qualified IBM Business Partners may place a full page
promotion in the most popular Redbooks publications.
Imagine the power of being seen by users who download
millions of Redbooks publications each year!
ibm.com/Redbooks
About Redbooks
Business Partner Programs
THIS PAGE INTENTIONALLY LEFT BLANK
Preface
IBM® SmartCloud Virtual Storage Center provides efficient virtualization and management of
heterogeneous storage systems. It facilitates migration to an agile cloud architecture that can
optimize storage availability and performance, while helping to reduce costs.
IBM SmartCloud® Virtual Storage Center (VSC) helps convert existing storage to IBM
Smarter Storage, providing more room for data growth and simplified storage administration.
This IBM Redbooks® publication gives an overview of the concepts of software-defined
environment (SDE) and software-defined storage (SDS), and how they work together with
VSC. It explores the architecture, components, and interfaces, providing details of VSC and
how to use it.
It also includes practical scenarios and use cases, helpful for client VSC business
environments, with a focus on the following topics:
򐂰 Introductory concepts: These can help you better understand SDE and SDS in both
cloud-based and traditional IT business environments.
򐂰 VSC components and available integrations: This topic describes how these can assist
you in transforming traditional storage to storage clouds.
򐂰 Storage management component of VSC: This topic shows you the functionality that the
integration of IBM Tivoli® Storage Productivity Center provides.
򐂰 Storage virtualization component of VSC: This topic gives an overview, comparisons, and
describes options with IBM SAN Volume Controller, Storwize® V5000 and V7000.
򐂰 Application aware data protection component of VSC: This topic gives an overview and
use case. Also provided are IBM Tivoli Storage FlashCopy® Manager and DB2® example
scripts (pre-freeze and post-thaw) and the steps for creating customer scripts before
running backups.
򐂰 VSC storage provisioning: This topic describes how you can simplify the way storage is
provisioned in an IT environment.
򐂰 VSC storage optimization: Functions and uses cases can help you to optimize your
storage environment by improving performance and to better utilize storage resources.
This book is primarily for storage administrators, users who are responsible for maintaining IT
and business infrastructures, and anyone who wants to learn more about IBM SmartCloud
Virtual Storage Center.
© Copyright IBM Corp. 2015. All rights reserved.
iii
Authors
This book was produced by a team of specialists from around the world working at the
International Technical Support Organization, Poughkeepsie Center.
Karen Orlando is a Project Leader at the International
Technical Support Organization, Tucson Arizona Center. Karen
has over 25 years in the IT industry with extensive experience
in open systems management, software development, and
hardware test of IBM storage. Karen has lead several projects
for Redbooks publications about IBM Tivoli Storage
Productivity Center, IBM FlashSystem™, IBM ProtecTIER®
products, and storage solutions. She holds a degree in
Business Information Systems from the University of Phoenix
and is Project Management Professional (PMP), certified since
2005.
Joseph Gatti is a Senior IT Specialist from Austin, Texas hired
by IBM in 1999. He has worked in IBM Software Group as a
Technical Enablement Specialist and in Advanced Technical
Presales and in Global Process Services as the lead IT
Architect and Storage Administrator. His areas of expertise are
IBM storage hardware and software, Brocade, Cisco, NetApp,
VMware vSphere, vCloud Director, Solarwinds Storage
Manager, and CloudStack. Joe has been providing pre- and
post-sales supported for Tivoli Storage Productivity Center
since version 1.2, and has created technical enablement
hands-on labs for FlashCopy Manager for VMware and Tivoli
Storage Manager for Virtual Environments.
Joao Marcos Leite is an IBM Senior Certified IT Specialist
from Brazil who joined IBM in 2000. For more than 13 years in
the IBM Systems and Technology Group, his focus has been in
the field of storage solutions design for clients. His current role
is in Client Technical Sales Support. His areas of expertise
include storage virtualization and storage management
products. Joao co-authored three previous Redbooks
publications for SAN Volume Controller, Tivoli Storage
Productivity Center, and SAN Volume Controller with
FlashSystem 820. He graduated as a Data Processing
Technologist from Universidade Federal do Parana in Curitiba,
Brazil, and has 33 years of experience as an IT Specialist. He
is a member of the Technology Leadership Council Brazil
(TLC-BR) an affiliate of the IBM Academy of Technology, and
holds a title of Distinguished IT Specialist by The Open Group.
iv
IBM SmartCloud Virtual Storage Center
Thirumalainambi Murugesh is an Integration Architect
working for the IBM Global Technology Services® (GTS) group
in Sydney, Australia. Murugesh has over 17 years of
experience in IT performing solution design, prototypes
development, implementation and integration. His areas of
expertise includes design and development of architectures,
patterns, assets and tools for high availability server and
storage solutions. Murugesh has designed infrastructures as a
service (IaaS) solution using VMware for servers virtualization,
IBM SmartCloud Virtual Storage Center. This includes SAN
Volume Controller for storage virtualization, Tivoli Storage
Productivity Center for storage management, Tivoli FlashCopy
Manager for application aware backups, Tivoli Storage
manager for data backups and Symantec Enterprise vault for
policy based data archiving. Murugesh has a Ph.D in Electrical
and Electronic Engineering from the University of Auckland,
New Zealand, with a specialty in Secure Network Management
with Cryptography.
Danijel Paulin is a Systems Architect at IBM in Croatia,
working for the Systems Architectural team, in the IBM South
East Europe region. He has 17 years of experience in IT.
Before joining IBM in 2003, he worked for two financial
companies in Croatia where he was responsible for IBM
mainframe and storage administration. He has experience with
a wide array of IBM storage products and storage software. His
areas of expertise include architecture and design of various
high availability, disaster recovery, business continuity, and
virtualization solutions for mainframe and open systems.
Socheat Sou is a Staff Software Engineer from Tucson,
Arizona. He has been with IBM in several capacities. He
worked in the storage technology area for six years, and with
the Tivoli Storage Productivity Center and Virtual Storage
Center for the past three years. He received a Computer
Science Engineering degree from the University of Michigan.
His areas of expertise include SONAS, OpenStack, and Linux.
Preface
v
Thanks to the following people for their contributions to this project:
Markus Standau
IBM Sales & Distribution, STG Sales
Thomas Clark
Brian De Guia
Scott McPeek
Bryan Odom
Harley Puckett
Ben Randall
IBM Software Group, Cloud & Smarter Infrastructure
Tony Pearson
IBM & Technology Group, Client Enablement & Systems Assurance, Master Inventor and
Senior Consulting IT Specialist
Toni Ionescu
Global Technology Services, Tivoli Storage Manager Client Product Currency Center of
Competency
Larry Coyne
IBM Sales & Distribution, Inside Sales, ITSO Tucson Center
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:
ibm.com/redbooks/residencies.html
vi
IBM SmartCloud Virtual Storage Center
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 found at:
ibm.com/redbooks
򐂰 Send your comments in an email to:
redbooks@us.ibm.com
򐂰 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
vii
viii
IBM SmartCloud Virtual Storage Center
1
Chapter 1.
Introductory concepts
This chapter provides the introductory concepts to help you better understand
software-defined environments (SDE) and software-defined storage (SDS) in both the
cloud-based and traditional IT business environments. We provide an overview of the IBM
SmartCloud Virtual Storage Center (VSC) component of this model, describing how this
solution enables your organization to deliver IT services in the most efficient way possible.
© Copyright IBM Corp. 2015. All rights reserved.
13
1.1 Overview
With the evolution of technologies, people through social media and smart devices are
generating huge amounts of data, which has lead to an information explosion in recent years.
In response, organizations are becoming more instrumented, interconnected, and intelligent.
While organizations continue to focus on improving and supporting their important traditional
IT business workloads like enterprise resource planning (ERP), human resources (HR),
customer relationship management (CRM), businesses are also striving to engage customers
that use big data, analytics, social media and mobile apps. Figure 1-1 depicts the results of
this information explosion.
Figure 1-1 The information explosion 1
The following sections provide an overview and the terminology to help you understand the
software-defined environment (SDE) model. The cloud-based model described in this book
encompasses IBM software-defined storage (SDS), with IBM SmartCloud Virtual Storage
Center (VSC).
We discuss in general how VSC is integrated into this environment, adds value, and delivers a
storage cloud-based solution.
1
Source: Tivoli Storage Talk, SmartCloud Virtual Storage Center (VSC) Sales Enablement
14
IBM SmartCloud Virtual Storage Center
1.2 Software-defined environment (SDE)
Imagine an entire IT infrastructure controlled not by hands and hardware, but by software:
one in which workloads like big data and analytics are serviced automatically by the most
appropriate resource.
This section briefly describes the SDE concept and the benefits provided to the business.
SDE can be described as a programmatic access to the infrastructure resources. The IT
industry is shifting from the traditional landscape to an application programming interface
(API) driven, cloud-based services consumption model. SDE transforms the old static IT
infrastructure resources into dynamic, constantly optimized, workload-aware, virtualized
resources.
SDE can be defined as a loosely coupled set of software components which are able to
provide, among others, the following basic functions in order to federate hardware resources
(such as compute, network, and storage) and make them all available in the form of an
integrated IT service:
򐂰 Virtualization and optimization
򐂰 Dynamic cross-domain integration and orchestration
򐂰 Policy-based workload automation
SDEs represent another transformational trend in the optimization journey to better support
today’s workloads (Figure 1-2).
Figure 1-2 SDE reference architecture
Chapter 1. Introductory concepts
15
The figure shows the following items:
Workloads
SDE improves agility of business applications and accelerates the
application lifecycle through rapid change. This layer represents the
traditional business applications (ERP, HR, CRM, and others) and the
new ones (big data, analytics, social, mobile, and others.). All of them
of course can have integration through APIs made available from the
API economy.
Service delivery
Cloud and traditional approaches to deliver the IT services to the
business applications can generally coexist. This layer represents both
as possible ways to claim IT resources from the SDE environment. In
cloud, for example, IT services can be requested through a web portal
or APIs and automatically generate a deployment workflow; in the
traditional approach they can be automated or even deployed
manually by operators.
IT infrastructure
This layer represents the IT SDE infrastructure for programmable,
open standards-based infrastructure foundation to enable cloud,
mobile and other dynamic enterprise solutions. This resource smart
infrastructure spans across all domains for maximum agility and
efficiency and consists of the following components:
Software-defined computing
Software-defined networking
Software-defined storage
1.2.1 IT SDE infrastructure components
Within the landscape of a software-defined environment, the IT Infrastructure layer spans
across all domain for the maximum agility and efficiency and consists of the software-defined
computing, software-defined networking, and the software-defined storage components.
Software-defined computing (SDC)
SDC is based on CPU virtualization technologies, which have been available in the market for
long time. Several hypervisor technologies exist, covering open and mainframe environments.
They are basically able to disconnect the physical hardware from the operating system,
setting up virtual machines (VMs) on the physical server to run business workloads.
In SDC environments, the value of optimization, and agility are achieved through a top-down
approach by these methods:
򐂰 Assessing the workload components
򐂰 Using patterns to determine the best ways to configure the components of middleware
applications
򐂰 Mapping the workload onto the available virtual resources
Software-defined networking (SDN)
SDN is a network paradigm that separates each network service from its point of attachment
to the network, creating a far more dynamic, flexible, automated, and manageable
architecture. Administrators can easily move virtual resources throughout the network, create
private virtual networks that meet specific performance security needs, and use many other
high-value applications. The key to SDN is an innovative approach to controlling how data
flows through a network. In a traditional network, data flow is controlled by switches and
routers.
16
IBM SmartCloud Virtual Storage Center
Each switch and router contains the following basic elements:
򐂰 Data plane: Physically carries data packets from one port to another by following rules that
are programmed into the device hardware. The data-forwarding plane operates at the
speed of the network (wire speed).
򐂰 Control plane: Contains the logic that the device uses to program the data plane, so
packets are forwarded correctly throughout the network.
򐂰 Management plane: An administrator can log in to the device and configure it for basic
activities. Most devices can be configured locally or through a network management tool.
Vendors use control plane software to optimize data flow to achieve high performance and
a competitive advantage. The switch-based control plane paradigm gives network
administrators little opportunity to increase data flow efficiency across the network as a
whole.
SDN abstracts the flow of control from individual devices to the network level. Similar to
server virtualization, where virtual machines are decoupled from the physical server,
network-wide virtualization gives administrators the power to define network flows that meet
the connectivity requirements of end stations and address the specific needs of discrete user
communities.
Software-defined storage (SDS)
SDS offerings unlock the potential of data and increase business agility and efficiency in ways
that were not possible until now. IBM software-defined storage enhances the speed and
efficiency of your storage and simplifies migration to new workloads by these methods:
򐂰 Accelerating global access to petabytes of data and billions of files
򐂰 Changing the economics of storage technology with multi-brand virtualization, intelligent
tiering, and open API support (OpenStack software, Apache Hadoop, and POSIX)
򐂰 Transforming business models with hybrid cloud storage and data protection for big data
analytics
򐂰 Supporting any storage, any data
The primary focus of subsequent topics in this chapter is on the SDS control plane layer, and
more specifically the value that VSC adds to cloud-based solutions from within this layer.
1.2.2 Software-defined storage overview
Software-defined storage (SDS) is a programmatic approach to storage. SDS is a set of
software capabilities that automatically manage data locally and globally, providing
breakthrough speed in data access and easier administration. It also provides the ability to
scale technology infrastructures quickly and more cost-effectively as data volumes expand. In
addition, these advances can work with any company’s storage systems to provide
automated and virtualized storage.
SDS consists of loosely coupled hardware and software components. They implement a
flexible, standard, and secure storage consumption model to programmatically meet your
workload requirements in real time. SDS provides SDE with the following capabilities:
򐂰
򐂰
򐂰
򐂰
Optimal workload allocation
Transaction integrity
Agility and scalability
Universal data access
Chapter 1. Introductory concepts
17
SDS inherent functions are as follows:
򐂰
򐂰
򐂰
򐂰
򐂰
򐂰
򐂰
Storage virtualization
Policy automation
Analytics and optimization
Backup and copy management
Integration and API services
Security
Cloud accessibility
The loosely coupled hardware and software products, with their specific features, of SDS
must be orchestrated in order to provide all the capabilities needed by an SDE. This model
means that customers do not need to replace their whole storage infrastructures with new
products or devices, but can save on their previous investments by integrating their current
storage systems into the new SDS environment.
The IBM SDS component model is illustrated in Figure 1-3. VSC is integrated in the IBM SDS
control plane. VSC incorporates advanced functions and storage management functions of
IBM Tivoli Storage Productivity Center, and also storage virtualization, and application-aware
data protection that is provided by Tivoli FlashCopy Manager advanced snapshot capabilities.
These components are coupled with advanced analytics, including storage optimization and
provisioning to provide a cloud-based solution.
Workload Optimized
Agility
Transaction Integrity
Rapid Scale
IBM SDS Control Plane
Integration &
API Services
Storage
Virtualization
Policy
Automation
Analytics &
Optimization
Backup & Copy
Management
Storage Cloud
Security
Open Common Control Plane
IBM SIS
SCSA
OpenStack APIs
VSC
TPC, FCM, TSM
VMWare
integration
Cloud service
portal
Cinder support
TPC, FCM
SVC Virtualization
Advanced Analytics
Enterprise Storage
management
IBM SDS Data Plane
Universal access
Inter-data center sharing
Efficient & secure storage
Cloud connection
File, object, POSIX, hadoop, iSCSI
Cross geo synchronization
Data ILM, encryption
Hybrid deployment
Scalable storage Software technologies
NAS / GPFS
Virtual Storage
Figure 1-3 SDS component model
18
IBM SmartCloud Virtual Storage Center
Object Storage
Block Storage
The figure shows two planes:
򐂰 SDS control plane: The control plane is a software layer that manages the virtualized
storage resources. It provides all the high-level functions needed by the customer to run
the business workload and enable an optimized, flexible, scalable and fast consumption of
the storage infrastructure. These capabilities span across several functions, such as
storage virtualization, policy automation, analytics and optimization, backup and copy
management, security, integration with the API services, and the possibility to use and
provide cloud storage services.
򐂰 SDS data plane: The data plane is the infrastructure where the data is processed. It
consists of the hardware infrastructure where the data is stored and all the basic storage
management functions such as RAID protection, tiering, copy services (remote, local,
synchronous, asynchronous, and point-in-time), encryption, and data deduplication, which
can be recalled and managed by the above layer. It provides a complete range of data
access possibilities, spanning from the traditional access methods like block I/O (for
example, iSCSI) or file I/O (for example, POSIX compliant), to the newer methods such as
object storage or the Hadoop Distributed File System (HDFS).
IBM SmartCloud Virtual Storage Center (VSC), as shown in Figure 1-3 on page 18, is a key
component of the IBM SDS control plane. It provides efficient virtualization and management
of heterogeneous storage systems. VSC provides storage virtualization, end-to-end storage
management, and data protection through application-aware snapshots. These capabilities
are all tightly integrated with advanced analytics functions, such as optimization and
provisioning, to deliver a robust storage, cloud-based, solution.
1.2.3 IBM SmartCloud Virtual Storage Center (VSC)
IBM SmartCloud Virtual Storage Center (VSC) provides an integrated storage management
environment that can centrally manage a heterogeneous storage environment. VSC can
automate a wide variety of storage management tasks (including virtualization and
provisioning) that enable storage to easily be integrated into a cloud computing environment.
IBM SmartCloud Virtual Storage Center V5.2 enables easy migration to agile, cloud-based
storage environments. This solution comprises the advanced functions available in IBM Tivoli
Storage Productivity Center, IBM System Storage® SAN Volume Controller, and IBM Tivoli
FlashCopy Manager, and the advanced analytics, provided by the IBM SmartCloud Virtual
Storage Center V5.2 license. As an end-to-end storage management solution included in
VSC, Tivoli Storage Productivity Center provides a single point of control that helps
administrators manage every aspect of the storage infrastructure, between the hosts, through
the fabric, and down to the physical disks for multi-site storage environments.
By integrating these products, VSC provides a combined storage virtualization platform and
storage management solution. It provides more room for data growth and simplifies the
administration of storage. Figure 1-4 on page 20 illustrates how VSC virtualization creates
synergy across the heterogeneous storage devices.
Chapter 1. Introductory concepts
19
Figure 1-4 VSC virtualization creates synergy across the heterogeneous storage devices
Subsequent chapters in this book offer details of VSC architecture, components, interfaces,
and practical scenarios that demonstrate use of VSC advanced analytics. Also included are
usage scenarios and discussions of storage provisioning and optimization.
1.3 The changing landscape of IT environments
Today more than ever, clients have a need to reduce their IT operational costs and
administration complexity to meet their modern business demands. To support this effort,
clients are adding to existing traditional IT environments and moving toward cloud computing,
offered in the market as commodity services.
The industry recognizes three primary models in cloud computing:
򐂰 Software as a service (SaaS):
–
–
–
–
Software runs on a cloud server.
Software is accessible through a web interface.
Users are concerned with only the front end (the web browser).
Examples are Google Apps, Dropbox, Evernote, and Skydrive.
򐂰 Platform as a service (PaaS):
– Is primarily for developers.
– Streamlines development, testing, and deployment.
– Examples are IBM Bluemix™, IBM managed Cloud Services, Google App Engine.
򐂰 Infrastructure as a service (IaaS):
–
–
–
–
Infrastructure and hardware are provided.
Model is pay-per-use.
Overhead is much lower.
Examples are IBM SoftLayer®, Rackspace, Amazon Elastic Cloud.
With cloud services available to consume for customers, traditional IT environments tend to or
will change over time toward a more software-defined structure. The potential savings in
capital expenditure (CAPEX) and operational expenditure (OPEX) motivate clients to change
the way they approach IT. Changing the way client IT environments work also involves
changing business processes and general concepts.
20
IBM SmartCloud Virtual Storage Center
Table 1-1 lists several types of traditional IT with new era IT applications and approaches.
Table 1-1 Traditional IT and new era IT
Traditional IT
New era IT
Banking, finance, insurance, mail
Web applications
ERP (SAP)
Media archives
Traditional Microsoft Office
Sync and share (Dropbox, Office365)
SQL Analytics
Hadoop Analytics
Transactional IT
Web scalability IT
1.4 Cloud computing
The National Institute of Standards and Technology (NIST) provides the following definition
for cloud computing2:
“Cloud computing is a model for enabling ubiquitous, convenient, on-demand network
access to a shared pool of configurable computing resources (e.g., networks, servers,
storage, applications, and services) that can be rapidly provisioned and released with
minimal management effort or service provider interaction.”
A company can use cloud services that are provided by third parties (public cloud). A
company can also build its own cloud (private cloud) and provide services from the cloud to
internal company users, eventually to selected business partners or customers, or to the
world. A company can also use both public and private (hybrid cloud).
A cloud service has the following characteristics:
򐂰
򐂰
򐂰
򐂰
򐂰
򐂰
Supports self-service provisioning
Accessibility through the Internet or corporate intranet
Provision resources from a pool, without the user needing any knowledge of it
Simple and fast resource elasticity, when users demand changes
Resources monitoring capability with a dashboard view of cloud health status
Metering capability, enabling a dynamic chargeback model
To provide these characteristics, the infrastructure that enables cloud services takes
advantage of two key enablers that, although not part of any formal cloud definition, are
proven to be indispensable in delivering the essential cloud service characteristics:
򐂰 Virtualization: Allows computing resources to be pooled and allocated on demand and
enables pay-per-use billing.
򐂰 Automation: Implements the elastic use of available resources, supporting the provisioning
and deprovisioning of service instances to support scalability, and allows the moving of
workloads where they are available.
Application programming interface (API)
The API plays a key role in software-defined environments. APIs consist of routines,
protocols, and tools to allow for software application communication. APIs specify how
software components should interact and be used. With an API, developers can use functions
of existing computer programs in other applications.
2
NIST Special Publication 800-145, P. Mell, T. Grance. The NIST Definition of Cloud Computing:
http://csrc.nist.gov/publications/nistpubs/800-145/SP800-145.pdf
Chapter 1. Introductory concepts
21
1.4.1 Storage cloud
Software-defined storage and storage cloud helps clients to address their top traditional
storage-related issues (“pain points”), represented in the data (from 2013 and 2014), as
shown in Figure 1-5.
Figure 1-5 Top storage-related issues (2013-2014 data)3
Storage for cloud
Storage for cloud is a general term that is applied to the type of storage environment that is
implemented in cloud computing to provision cloud-computing services. For example, when a
virtual server machine is created, some storage capacity is required. This storage is
provisioned as part of the VM creation process to support the operating system and runtime
environment for the instance. It is not delivered by a storage cloud. However, it can be
provisioned from the same storage infrastructure as a storage cloud.
A storage cloud can be part of an overall cloud data center, or can provide storage as a
service to storage users. It can be delivered in any of the cloud delivery models (public,
private, hybrid, or for example the OpenStack community, which is described in 2.5.1,
“OpenStack software” on page 41). A storage cloud can be used to support a diverse range
of storage needs, including mass data stores, file shares, backup, archive, and more. Several
examples of publicly available storage clouds are as follows:
򐂰 IBM SmartCloud solutions, which offers various storage options, including archive,
backup, and object storage.
򐂰 Microsoft OneDrive, which can be used to store and share files on the Microsoft public
storage cloud service.
򐂰 Email services such as Gmail and Yahoo, which are services that store user email and
attachments in their storage clouds.
򐂰 Facebook and YouTube, which allow users to store and share photos and videos.
3
22
Source is from InfoPro: Storage Study: Wave 18 - 451 Research/The InfoPro
IBM SmartCloud Virtual Storage Center
Storage cloud capability can also be offered in the form of storage as a service, where fees
are paid based on the amount of storage space that is used. There are various ways a
storage cloud can be used, based on your organization’s specific requirements.
1.4.2 Benefits of storage cloud
The storage cloud model can be helpful for businesses that have seasonal or unpredictable
capacity requirements, and for businesses that require rapid deployment or contraction of
storage capacity.
When storage cloud is used, clients can focus on their core business, and worry less about
supporting a storage infrastructure for their data. Storage cloud provides the following key
advantages:
򐂰 Facilitates rapid capacity provisioning, supporting business agility
򐂰 Improves storage usage by avoiding unused capacity
򐂰 Provides support for storage consolidation and storage virtualization functionality
򐂰 Gives chargeback and showback accounting for usage as an incentive to conserve
resources.
Can be used to manage IT infrastructure and is based on the user’s needs.
򐂰 Offers integrated storage and service management
The storage cloud infrastructure usually includes integrated management software. This
software can be used to help manage the complete storage infrastructure from a single
console, without having to buy proprietary management software from multiple vendors.
This method can save time and help reduce spending on various proprietary management
software.
򐂰 Improves efficiency of data management
Consolidation and standardization of storage resources facilitates less infrastructure
complexity, which simplifies storage management. Consistent policies and process with
integrated management tools support geographically diverse infrastructure requirements
that are driven by performance or availability considerations.
򐂰 Provides faster time to market
Automation, self-service portals, rapid deployment, dynamic scaling, and centralized
storage management enhance business agility by facilitating significant improvements,
such as decreased time-to-market for new products. Businesses can focus on building
their core products and competencies rather than worrying about the management of their
IT infrastructure.
1.4.3 IBM SmartCloud Virtual Storage Center and storage cloud
IBM SmartCloud Virtual Storage Center (VSC) is the cornerstone of SDS control plane and it
provides several capabilities in forming a storage cloud solution. According to the IBM SDS
model, the control plane provides the storage management services that can be used by
SDEs (either VMware based or Openstack software based), a series of integration APIs
(based on REST and Open Service for Lifecycle Collaboration (OSLC) and the possibility of
storage provisioning. SDS can also be integrated as an IT infrastructure component of cloud
service providers' environments like SoftLayer, or vice versa it can exploit storage resources
available in the storage cloud.
Chapter 1. Introductory concepts
23
If you answer “yes” to any of the following questions, consider storage cloud for your storage
environment.
򐂰 Do you have a need to consolidate mixed-vendor storage devices under a single point of
control?
򐂰 Is determining how much storage is available for your applications difficult? What about
how the storage is being used?
򐂰 Is storage affecting availability of your applications?
򐂰 Do you find that limitations imposed by your storage are controlling how you deploy new
applications?
򐂰 Are you concerned with how fast your storage environment is growing? Are you concerned
with how well-utilized your storage is?
򐂰 Is forecasting and planning for storage purchases difficult?
򐂰 Do you need to reduce storage acquisition and storage management costs while at the
same time increase availability of your storage resources?
򐂰 Do you lack tools for analytics, trending, problem determination, and asset reporting that
enable you to maximize the reliability of your SAN?
򐂰 Would you like a way to dynamically identify hotspots on your storage systems and then
migrate data from one location to another location without any disruption to your service
level and users?
򐂰 Are you concerned about the amount of time required by administration staff to manage
your growing storage?
Figure 1-6 shows how the SDS control plane can be implemented by IBM SmartCloud Virtual
Storage Center.
Figure 1-6 IBM SmartCloud VSC in SDS control plane to form a storage cloud solution
24
IBM SmartCloud Virtual Storage Center
IBM VSC offers the following features in forming a storage cloud solution:
򐂰 Virtualization: VSC virtualizes heterogeneous storage at a block I/O level regardless of the
vendors.
򐂰 Optimization: VSC optimizes the use of the block storage, creating a unique physical pool
from which the customers can create their own tiers of storage.
򐂰 Analytics: VSC provides advanced storage analytics capabilities to discover, monitor, and
report about the disk capacity usage and performances, and implement the chargeback; it
simplifies the root cause analysis of performance issues.
򐂰 Policy automation: With VSC, customers can implement standardized storage provisioning
by automation, with policies.
򐂰 Copy Management: Clients can use VSC to produce and manage multiple point-in-time
copies of the data for backup, which can be sent to Tivoli Storage Manger for backup to
media.
򐂰 Improved storage utilization: Make better use of existing storage and controlling storage
growth expenditures.
򐂰 Improved application availability and simplified data migrations: Clients can make changes
to storage and move data without taking applications down.
򐂰 Simplified storage management: VSC offers greater efficiency and productivity for storage
management staff.
򐂰 Network-based replication and software architectural advantages: Helps enable greater
choice (lower cost) when buying storage and lowers software costs.
򐂰 Reduction in storage management and administration cost: A core group of administrators
can control multiple assets across a distributed storage environment (efficiency is
improved by 50%).
򐂰 Improved storage utilization: Improves capacity utilization of existing storage assets.
򐂰 Controls future spending: Controls the growth of future spending (improves use by 30%).
򐂰 Reduced cost of storage: Capitalize on being able to purchase the lowest cost storage
resources (growth is controlled, on average, by 20%).
򐂰 Improved customer and user availability to data-driven applications: Minimizes downtime
associated with migrating data between storage assets
IBM SmartCloud Virtual Storage Center components and integration is described in
Chapter 2, “Virtual Storage Center (VSC)” on page 27.
Chapter 1. Introductory concepts
25
26
IBM SmartCloud Virtual Storage Center
2
Chapter 2.
Virtual Storage Center (VSC)
This chapter describes the components of IBM SmartCloud Virtual Storage Center (VSC). At
a high level, this chapter describes features of the components and available integrations to
help you in transforming traditional storage to storage cloud.
Also described in this chapter are the VSC offerings and licensing model overview.
© Copyright IBM Corp. 2015. All rights reserved.
27
2.1 IBM SmartCloud Virtual Storage Center overview
IBM SmartCloud Virtual Storage Center (VSC) 5.2 provides efficient virtualization,
management and data protection for heterogeneous storage environments. VSC helps IT
storage managers migrate to an agile cloud-based storage environment and manage it
effectively without having to replace existing storage systems. This powerful offering removes
the physicality of storage, and also the complexity that is associated with managing
multivendor infrastructures.
VSC V5.2 offers a storage virtualization platform, capabilities for storage virtualization
management, and instant copy management. VSC V5.2 delivers to customers, under one
licensed software product, the complete set of functions available in the IBM Tivoli Storage
Productivity Center, the functions and capabilities that are associated with the IBM System
Storage SAN Volume Controller (including copy services), and the capabilities of the IBM
Tivoli Storage FlashCopy Manager. With VSC, you can now get all of the advanced
capabilities of what was previously Tivoli Storage Productivity Center Standard Edition; and
with the IBM SmartCloud VSC 5.2 license you get all of the advanced analytics functions.
This powerful solution enables organizations to optimize provisioning, capacity, availability,
reporting, and management for virtual storage.
2.2 Storage hypervisor
This topic introduces server hypervisor and storage hypervisor. It also has an overview of the
IBM Storage Hypervisor, which is integrated with the IBM SmartCloud Virtual Storage Center
V5.2.
Server hypervisor
In cloud computing, server hypervisor is a well understood term, with the following key
attributes, in providing effective resource utilization, cost savings, and flexibility to the
business:
򐂰 Pooled physical resources are consumed by virtual machines resulting in high asset
utilization.
򐂰 Virtual machines are mobile, giving administrators their choice of physical server and
location.
򐂰 A common set of value capabilities and centralized management are provided for virtual
machines, regardless of what physical server they are running on.
Storage hypervisor
Storage hypervisor is a rapidly emerging way of describing the same value aspects, but in a
storage context:
򐂰 Consolidation and cost: Storage pooling increases utilization and decrease costs.
򐂰 Business availability: Data mobility of virtual volumes can improve availability.
򐂰 Application support: Tiered storage optimization aligns storage costs with required
application service levels.
28
IBM SmartCloud Virtual Storage Center
IBM Storage Hypervisor
The IBM Storage Hypervisor offers the following features (shown in Figure 2-1):
򐂰 Virtualizes storage resources from multiple arrays, vendors, and data centers, which are
pooled together and accessed anywhere.
򐂰 Standardized storage services are selected from a service catalog.
򐂰 Mobility of storage volumes move dynamically based on workload balancing policies.
򐂰 Self-service provisioning uses automation to allocate capacity.
򐂰 Pay-per-use storage resources so users are aware of the impact of their consumption,
service level choices.
Figure 2-1 IBM Storage Hypervisor
IBM Storage Hypervisor is part of the IBM SmartCloud Virtual Storage Center 5.2 (VSC)
which includes storage virtualization, storage virtualization management, and storage
snapshot management that are tightly integrated with advanced analytics to deliver a robust
storage cloud solution. This solution ultimately helps businesses to optimize provisioning,
capacity, availability, data protection, reporting, and management for virtualized storage.
Chapter 2. Virtual Storage Center (VSC)
29
2.3 IBM SmartCloud Virtual Storage Center component model
As shown in Figure 2-2, IBM SmartCloud Virtual Storage Center 5.2 includes core
functionality, from three IBM offerings; Storage management through IBM Tivoli Storage
Productivity Center, storage virtualization with IBM System Storage SAN Volume Controller,
and application-aware data protection with IBM Tivoli Storage FlashCopy Manager.
Figure 2-2 Overview diagram of IBM SmartCloud Virtual Storage Center
2.3.1 Storage management
The storage management component in IBM SmartCloud VSC V5.2 provides an advanced
storage infrastructure and data management capabilities. The Tivoli Storage Productivity
Center component available in VSC includes all the capabilities of Tivoli Storage Productivity
Center V5.2. It uniquely provides all the advanced functions that were available in the past as
part of Tivoli Storage Productivity Center Standard Edition and Tivoli Storage Productivity
Center for Replication. Unique to the VSC V5.2 Storage Analytics Engine is data
management with file system and database scanning and analysis, data placement, user
quotas, and an advanced management GUI to help simplify virtual storage administration.
The storage management component of the VSC solution is designed to improve visibility,
control, and automation for data and storage infrastructures, including storage systems,
devices, and SAN fabrics and integrated with SAN Volume Controller functionality for
auto-tiering and workload-aware placement across the data center.
30
IBM SmartCloud Virtual Storage Center
Tivoli Storage Productivity Center, the storage management component of VSC, helps
simplify provisioning, performance management, and data replication processes (Figure 2-3).
Figure 2-3 VSC storage management and Tivoli Storage Productivity Center
IBM Tivoli Storage Productivity Center provides these capabilities, all from a single GUI:
򐂰 Database, host, file system, and file level capacity analytics
򐂰 Storage performance management
򐂰 Tiered storage analysis
򐂰 Trend analysis
򐂰 SAN planning and provisioning
򐂰 Performance optimization
򐂰 SAN fabric performance management
Note: For details about VSC offerings and licensing, see 2.6, “IBM Virtual Storage Center
offerings” on page 51.
Chapter 2. Virtual Storage Center (VSC)
31
Tivoli Storage Productivity Center can generate threshold alerts and forward them to SNMP
Receivers. Tivoli Storage Productivity Center provides many ready-to-use reports as shown
in Figure 2-4.
Figure 2-4 Tivoli Storage Productivity Center reports that are ready to use
These reports can be scheduled to run periodically. Additional custom reports can be created
with IBM Cognos®.
For more information about IBM Cognos reports, see the “Enhanced reporting with IBM
Cognos” topic in IBM Tivoli Storage Productivity Center V5.1 Technical Guide, SG24-8053.
Additional capabilities of Tivoli Storage Productivity Center are described in Chapter 3,
“Storage management” on page 57.
32
IBM SmartCloud Virtual Storage Center
2.3.2 Storage virtualization
The IBM SAN Volume Controller virtualization engine moves the storage control function into
the storage network, allowing disk storage to be managed as a single virtual pool, supporting
a large number of disk vendors (Figure 2-5).
Figure 2-5 SAN Volume Controller conceptual and topology overview
SAN Volume Controller can pool storage volumes into a reservoir of capacity for centralized
management. Virtualization with SAN Volume Controller eliminates the boundaries among
disk and flash systems, which simplifies management and enables IT Operations to focus on
managing storage as a resource to meet business requirements rather than as a set of boxes.
The RAID array from an external storage system or from internal disks (Storwize V7000, as
shown in our example in Figure 2-5) is presented to an SAN Volume Controller or Storwize
V7000 as Managed Disks or MDisks. A set of MDisks forms a storage pool from which
extents are taken to create the volumes, which can be identified by logical unit numbers
(LUNs). The volumes, now in virtualized mode, are presented to the hosts. In this sense, the
hosts no longer see the back-end disks directly, and the SAN Volume Controller or Storwize
V7000 behaves like a controller that is provisioning LUNs to the hosts.
Chapter 2. Virtual Storage Center (VSC)
33
To achieve multitenancy over the same physical SAN infrastructure, storage pools can be
created that are specific to each tenant from a specific set of managed disks and assign them
to the specific tenant hosts, as depicted in Figure 2-6.
Figure 2-6 SAN Volume Controller storage virtualization concepts summary
The SAN Volume Controller component of VSC reduces labor, reduces and removes planned
migration outages, and improves utilization. Storage virtualization with IBM SAN Volume
Controller supports a heterogeneous, multivendor environment, with common management
and services. SAN Volume Controller allows for nondisruptive changes to the storage
environment without impacting host applications. SAN Volume Controller with Infrastructure
Lifecycle Management (ILM) intelligent storage analytics provides policy-based automated
data placement and tier movement.
The key characteristics of SAN Volume Controller are as follows:
򐂰 Highly scalable: A SAN Volume Controller cluster scales horizontally through the addition
of node pairs to a maximum of four node pairs (or eight nodes) per cluster.
򐂰 Host-independent: Supports multiple operating systems Windows, Linux, IBM AIX®,
HP-UX and so on.
򐂰 Storage controller-independent: Supports storage from multiple vendors including IBM,
EMC, HDS, Oracle, Hewlett-Packard, and others.
34
IBM SmartCloud Virtual Storage Center
SAN Volume Controller offers the following services:
򐂰 Creation and management of a storage pools attached to the SAN.
򐂰 Block level virtualization.
򐂰 Provision of advanced functions across the SAN such as advanced copy services
(point-in-time copy, instant copy, synchronous remote copy, Metro Mirror and
asynchronous remote copy, Global Mirror).
򐂰 Thin provisioning.
򐂰 Real-time compression: The IBM Real-time Compression™ option can be added as a
separate priced license. For more details about this topic see 2.6, “IBM Virtual Storage
Center offerings” on page 51.
򐂰 Data migration: Move volumes within or between storage controllers (within the same
physical virtualization boundary).
򐂰 Growing or shrinking volumes.
򐂰 IBM Easy Tier® helps administrators control storage growth more effectively by balancing
MDisks within a pool, and by moving low-activity or inactive data into a hierarchy of
lower-cost storage. Administrators can free disk space on higher-value storage for more
important, active data.
For more details about SAN Volume Controller Easy Tier functionality, see the following
publications:
– Implementing IBM FlashSystem 840, SG24-8189 (see Chapter 8, “Product integration”
and Chapter 9, “Use cases and solutions”)
– IBM SAN Volume Controller 2145-DH8 Introduction and Implementation, SG24-8229
(see Chapter 5, “SAN Volume Controller Easy Tier”)
– IBM System Storage SAN Volume Controller Best Practices and Performance
Guidelines, SG24-7521 (see Chapter 11, “IBM System Storage Easy Tier function”)
– IBM DS8000 Easy Tier, REDP-4667
Additional capabilities of SAN Volume Controller and the IBM Storwize product family are
discussed in Chapter 4, “Storage virtualization” on page 89.
Chapter 2. Virtual Storage Center (VSC)
35
2.3.3 Application-aware data protection
With the Tivoli Storage FlashCopy Manager component of VSC, the data backup and restore
component in IBM SmartCloud VSC V5.2 provides fast application-aware backups and
restores, leveraging advanced snapshot technologies available with IBM storage systems
(Figure 2-7).
Figure 2-7 High level overview of Tivoli Storage FlashCopy Manager application-aware copies solution
FlashCopy Manager uses advanced IBM storage hardware snapshot technology to help
create a high performance, low impact application data protection solution. The Storage
FlashCopy function operates at the level of virtual volumes (VDisks). That is, it copies whole
volumes. The FlashCopy function is intended to create copies of data that may be used for
purposes such as disk-to-disk backups, parallel processing (multiple applications processing
different copies of the same data), testing by using a copy of production data, and so on. The
copy created by the FlashCopy function is available to be used almost immediately. SAN
Volume Controller can perform a background copy of all data from the source to the target or
it can copy data only when an update occurs. It delivers high levels of data protection for
business-critical applications through integrated application snapshot backup and restore
capabilities.
Storage administrators can control the speed of the background copy to limit the impact that
the copy has on other SAN Volume Controller activities. The first time that FlashCopy is used,
the copy takes place as normal, which means that a full copy of data occurs from the source
to the target VDisk. When changes are made, only the changes are copied to the target. A
total of 256 copies can be made from the same source VDisk, which can be incremental or
non-incremental, or a mix of both.
36
IBM SmartCloud Virtual Storage Center
It integrates with IBM System Storage DS8000®, FlashSystem V840, IBM SAN Volume
Controller, IBM Storwize V7000 and V5000, and IBM XIV® Storage System products. For
Microsoft Windows environments, Tivoli Storage FlashCopy Manager also supports other
hardware that is capable of Microsoft Volume Shadow Copy Services (VSS) functions, such
as IBM N Series, System Storage DS3000, DS4000®, and DS5000™.
The FlashCopy Manager solution is explained (see Figure 2-7 on page 36) as follows:
1. Starting at the left, the Application system, which is also called the Production system, has
the production database on it. More important, this is the data that you want to protect.
The applications can be Oracle, SQL, DB2, SAP, Exchange, Files Systems, or VMware.
FlashCopy Manager also provides the interfaces for custom applications to take
snapshots the data.
2. Following the black arrow, you see that the application data resides on the LUN that sits on
the SAN Volume Controller and on its back-end storage. Using FlashCopy Manager, when
you take the backup of the database, you have local snapshot versions that represent the
application data at some point in time. When you want to restore the data, use FlashCopy
Manager to restore from any one of these snapshot versions including the latest to a
point-in-time snapshot. If you have Tivoli Storage Manager. You can then offload your
backups to Tivoli Storage Manager and manage your data through that and FlashCopy
Manager.
Additional capabilities of FlashCopy Manager are discussed in Chapter 5, “Application-aware
data protection” on page 103.
2.4 IBM SmartCloud Virtual Storage Center features
IBM SmartCloud Virtual Storage Center helps reduce storage administration complexity and
costs by these ways:
򐂰
򐂰
򐂰
򐂰
򐂰
򐂰
򐂰
򐂰
򐂰
򐂰
򐂰
Improving storage utilization
Making better use of existing storage and controlling storage growth expenditures
Improving application availability and simplified data migrations
Making changes to storage and move data without taking applications down
Simplifying storage management
Improving efficiency and productivity for storage management staff
Providing advantages with software-defined storage architecture model
Enabling greater choice (lower cost) when buying storage and lowers software costs
Improving application recovery time and recovery point objectives (RTO and RPO)
Providing application-aware hardware-based snapshots
Providing network-based replication
The outstanding features of IBM SAN Volume Controller are as follows:
򐂰 Efficient by design
򐂰 Self-optimizing
򐂰 Cloud agility
2.4.1 Efficient by design
Organizations need to spend less of their IT budgets on storage and storage administration
so that they can spend more on new, revenue-generating initiatives. IBM SmartCloud Virtual
Storage Center has built-in efficiency features that help users avoid purchasing add-ons or
additional licenses or deal with complicated integration issues.
Chapter 2. Virtual Storage Center (VSC)
37
It has these advanced efficiency features:
򐂰 Storage virtualization
This is a foundational technology for clouds and software-defined environments. Without
virtualization, storage capacity utilization averages about 50%, but virtualized storage
enables up to 90% utilization by enabling online data migration for load balancing. With
IBM SmartCloud Virtual Storage Center, you can virtualize your storage resources from
multiple storage systems and vendors. Pooling storage devices enables you to access
capacity from any storage system, a significant advantage over the limitations inherent in
traditional storage.
򐂰 Simplified user experience
IBM SmartCloud Virtual Storage Center provides an advanced GUI and a VMware
vCenter plug-in to reduce administration complexity. Administrators can do common tasks
consistently, over multiple storage systems, even those from different vendors. The IBM
storage GUI enables simplified storage provisioning with intelligent presets and embedded
best practices, and integrates context-sensitive performance management throughout.
򐂰 Near-instant, application-aware backup and restore
To reduce downtime in high-availability virtual environments, critical applications such as
email and databases require near-instant backups that have little or no impact on
application performance. Application-aware snapshot backups can be performed
frequently throughout the day to reduce the risk of data loss. IBM SmartCloud Virtual
Storage Center simplifies administration and recovery from snapshot backups.
2.4.2 Self-optimizing
Self-optimizing storage adapts automatically to workload changes to optimize application
performance, eliminating most manual tuning efforts. IBM SmartCloud VSC includes these
self-optimizing features:
򐂰 IBM Tiered Storage Optimizer
IBM SmartCloud Virtual Storage Center uses performance metrics, advanced analytics,
and automation to enable storage optimization on a large scale. It can optimize storage
volumes across different storage systems and virtual machine vendors and brands. The
Tiered Storage Optimizer feature can reduce the unit cost of storage by as much as 50%,
based on deployment in a large IBM data center.
򐂰 IBM Easy Tier
IBM SmartCloud Virtual Storage Center helps optimize flash storage with automated
tiering for critical workloads. Easy Tier helps make the best use of available storage
resources by automatically moving the most active data to the fastest storage tier, which
helps applications and virtual desktop environments run up to three times faster.
򐂰 Thin provisioning and efficient remote mirroring
Thin provisioning helps automate provisioning and improve productivity by enabling
administrators to focus on overall storage deployment and utilization, and also on longer
term strategic requirements without being distracted by routine storage-provisioning
requests. IBM Metro Mirror and Global Mirror functions automatically copy data to remote
sites as it changes, enabling fast failover and recovery. These capabilities are integrated
into the advanced GUI, so that they become easier to deploy.
38
IBM SmartCloud Virtual Storage Center
2.4.3 Cloud agility
Cloud computing is all about agility. Storage for clouds must be as flexible and
service-oriented as the applications it supports. IBM SmartCloud Virtual Storage Center can
convert existing storage into a private storage cloud with no “rip and replace” required. The
solution enables you to adapt to the dynamic storage needs of cloud applications by providing
storage virtualization, automation, and integration for cloud environments.
Features are as follows:
򐂰 OpenStack cloud application provisioning
IBM SmartCloud Virtual Storage Center includes an OpenStack Cinder volume driver that
enables automated provisioning using any of the storage systems controlled by IBM
SmartCloud Virtual Storage Center. OpenStack cloud applications can access multiple
storage tiers and services, without added complexity.
򐂰 Self-service portal
IBM SmartCloud Virtual Storage Center can provide provisioning automation for
self-service storage portals (such as IBM SmartCloud Storage Access), which enable
immediate responses to service requests while eliminating manual administration tasks.
򐂰 Pay-per-use invoicing
IBM SmartCloud Virtual Storage Center integrates with IBM SmartCloud Cost Manager
and other chargeback systems to enable flexible usage accounting for storage resources.
IBM SmartCloud Virtual Storage Center can become the single source for usage metrics
across storage area networks (SANs), network-attached storage, and direct-attached
storage.
Chapter 2. Virtual Storage Center (VSC)
39
2.5 IBM SmartCloud Virtual Storage Center interfaces
IBM focuses on supporting four software-defined environments (SDE), shown in Figure 2-8.
Figure 2-8 Interfaces to IBM SmartCloud Virtual Storage Center
The four SDEs are as follows:
򐂰
򐂰
򐂰
򐂰
40
OpenStack open source code can manage compute, network and storage resources.
IBM SmartCloud is based on OpenStack with value-added IBM proprietary features.
IBM Cloud Orchestrator is for storage provisioning, orchestration. and automation.
VMware runs primarily on x86-based servers.
IBM SmartCloud Virtual Storage Center
The interface features are summarized in Table 2-1.
Table 2-1 Comparison of virtual storage interfaces to software-defined environment
OpenStack software
IBM SmartCloud and
IBM Orchestrator
VMware
IBM is a platinum sponsor of
OpenStack Foundation.
IBM Cloud Manager with
OpenStack is based on
OpenStack open source code,
with value-added proprietary
features from IBM.
VMware is entirely proprietary,
but has large market share for
x86-based server
infrastructure.
OpenStack open source code
can manage IBM compute,
network and storage resources.
IBM Cloud Manager with
OpenStack and IBM Cloud
Orchestrator support various
server hypervisors and
interfaces.
IBM was VMware's first OEM
and joint development partner
(since 1998). IBM continues
strong partnership. IBM Global
Services is one of VMware's
largest customers, using
VMware in many of their client
solutions.
IBM offers Cinder interfaces on
most of its major storage
products for Block storage
access and supports Swift
interfaces for object storage
access.
IBM SmartCloud Storage
Access and IBM Cloud
Orchestrator provide
self-provisioning and
orchestration capabilities.
VMware vStorage API for data
protection (VADP), VMware
Site Recovery Manager (SRM),
VMware vSphere storage APIs:
Array integration (VAAI),
VMware vCenter.
An overview of each VSC interface is provided in the following sections.
2.5.1 OpenStack software
The OpenStack environment is a cloud operating system that controls large pools of
compute, storage, and networking resources throughout a data center, all managed through a
dashboard that gives administrators control while empowering users to provision resources
through a web interface. OpenStack Icehouse is the ninth release of the open source
software for building public, private, and hybrid clouds, has nearly 350 new features to
support software development, managing data and application infrastructure at scale.
The OpenStack community continues to attract the best developers and experts in their
disciplines with over 1,200 individuals employed by more than 120 organizations contributing
to the Icehouse release.
Because OpenStack design and development is done in the open, public documentation is
available regarding the development status of the current release and decisions made at each
Design Summit. To review details of this information, see the Icehouse Release Notes:
https://wiki.openstack.org/wiki/ReleaseNotes/Icehouse
OpenStack Cinder support
The Tivoli Storage Productivity Center OpenStack Cinder driver enables your
OpenStack-powered cloud to use your Tivoli Storage Productivity Center installation for block
storage provisioning.
Tivoli Storage Productivity Center provides block storage provisioning capabilities that the
storage administrator can employ to define the properties and characteristics of storage
Chapter 2. Virtual Storage Center (VSC)
41
volumes within a particular service class. For example, a block storage service class may
define RAID levels, tiers of storage, and various other storage characteristics.
The Tivoli Storage Productivity Center Cinder driver enables your OpenStack cloud to use
these defined service classes, which are exposed directly into your Cinder environment as
Cinder volume types. For each block storage service class, a Cinder volume type is created.
This enables your cloud users to create and use volumes of this type.
Figure 2-9 shows the interaction between the OpenStack components (Horizon, Nova, and
Cinder) and Tivoli Storage Productivity Center. When an OpenStack cloud user requests a
new volume for a virtual machine through the Horizon dashboard, the Cinder node sends
Tivoli Storage Productivity Center all the necessary information through the Cinder driver,
which includes service class information. With that information, Tivoli Storage Productivity
Center chooses the most appropriate storage system to use and creates a volume and all the
necessary connections. Tivoli Storage Productivity Center then passes that information back
to the Cinder node through the same driver. From there, the Cinder node assigns the volume
to the virtual machine.
Figure 2-9 Tivoli Storage Productivity Center Cinder driver components
2.5.2 IBM SmartCloud Storage Access
IBM SmartCloud Storage Access is a software appliance that provides cloud services to block
and file storage environments. IBM SmartCloud Storage Access Version 1.2 or later supports
the following storage devices:
򐂰
򐂰
򐂰
򐂰
򐂰
򐂰
򐂰
IBM Scale Out Network Attached Storage (SONAS)
IBM Storwize V7000 Unified
IBM Storwize V7000
IBM Storwize V5000
IBM FlashSystem V840
IBM System Storage SAN Volume Controller
IBM XIV Storage System
IBM SmartCloud Storage Access provides an environment where the users can serve
themselves. The cloud administrators, through IBM SmartCloud Storage Access, can define
polices and automate tasks for provisioning resources.
42
IBM SmartCloud Virtual Storage Center
The following list includes some of the key storage cloud operations that are enabled through
IBM SmartCloud Storage Access:
򐂰 Creating and managing file storage and Fibre Channel block storage
򐂰 Creating working environments to give team-specific access to file and block storage
򐂰 Providing request and approval work flow support
򐂰 Providing access to file space through Network File System (NFS) and Common Internet
File System (CIFS)
򐂰 Setting automatic mail notifications according to policy
򐂰 Providing support for expiration of storage resources
򐂰 Providing support to federated authentication
򐂰 Providing a Storage Cloud API for third-party integration
򐂰 Providing globalization support
IBM SmartCloud Storage Access deployment with VSC architecture
This section describes a deployment architecture for a private storage cloud with IBM
SmartCloud Storage Access. As shown in Figure 2-10 on page 44, IBM SmartCloud Storage
Access works with various external systems or services to achieve the functionality required
for a private storage cloud.
Chapter 2. Virtual Storage Center (VSC)
43
Figure 2-10 IBM SmartCloud Storage Access deployment architecture diagram
Three words that matter the most to any enterprise are reduce, reuse, and recycle. These are
true for private cloud deployment based on IBM SmartCloud Storage Access, because the
enterprises can achieve the following goals:
򐂰 Reduce administrator resource usage, operational costs, and provisioning turnaround
time.
򐂰 Reuse existing infrastructure components, such as NTP server, DNS, SMTP server, and
storage systems.
򐂰 Recycle unused or underused storage space, and therefore ensure optimal use.
The deployment architecture as shown in Figure 2-10, has IBM SmartCloud Storage Access
working with Tivoli Storage Productivity Center, which is the storage manager to provision
and deprovision network drives. Tivoli Storage Productivity Center plays a major role in
understanding the resource request from IBM SmartCloud Storage Access and completing
the resource provisioning.
44
IBM SmartCloud Virtual Storage Center
2.5.3 IBM Cloud Orchestrator
IBM Cloud Orchestrator is built on a foundation of open standards, including OpenStack and
community-provided best practices, and provides common cloud services for compute,
storage, and network resources, while also supporting multiple hypervisors and multivendor
platforms. It is designed to simplify and speed the creation of process workflows that can help
shorten deployment and change processes in the cloud.
With IBM Cloud Orchestrator, cloud administrators can use a single interface to design and
deploy services, monitor capacity and performance, control updates and migrations, and
recover unused resources as needed. It also helps IT staff deliver resources to users more
quickly and consistently.
Managing workloads across the lifecycle requires orchestration. IBM Cloud Orchestrator
performs Resource Orchestration, Workload Orchestration, and Service Orchestration, as
shown in Figure 2-11). A Cloud Orchestrator administrator can create a process that will
onboard a VM, provision and manage the CPU, storage and network, and place the new
system based on existing workloads. The process can optimize the performance and
operation of the server. Finally, the entire lifecycle of the business application is managed,
ensuring that patches are applied in a timely manner and that service tickets are opened
when required.
Figure 2-11 IBM Cloud Orchestrator
Assets are built, monitored, and managed from a single product. Individual software titles or
bundles are available to extend the base virtual image as necessary. Patterns are available
for use to deploy complex applications or environments and may generate entire systems. For
example, a pattern might include a web component on one system, a database on a second
system and a file server on a third. The number and type of systems that can be built using a
pattern is determined by the software bundles that are available and the number of resources
available.
Chapter 2. Virtual Storage Center (VSC)
45
The orchestration processes tie in the customizations that are required for environments,
including specific user interfaces (perhaps a custom web front end), user interface widgets,
and the business processes that handle the email notification to requesters, and the defined
approval processes. The orchestration component can also perform the steps to
decommission the virtual machine or environment either manually or based on an end date
that is provided at the time the virtual machine or environment was requested.
In summary IBM Cloud Orchestrator provides the following features:
򐂰 Standardizes and automates cloud services through a flexible orchestration engine and a
self-service portal.
򐂰 Provides reusable workload patterns to enable dynamic cloud service delivery.
򐂰 Is built on open standards, including OpenStack cloud software, for unparalleled
interoperability.
򐂰 Use of open technologies such as OpenStack cloud software to build an interoperable
infrastructure foundation to provision workloads, provide multitenancy, and enable
administration.
򐂰 Offers infrastructure and workload orchestration for VMware and Microsoft virtualization
infrastructures, and adopts OpenStack technology.
򐂰 Supports deployments on public clouds such as Amazon Elastic Compute Cloud, allowing
the implementation of a hybrid cloud model.
Standardization and automation of cloud services with IBM Cloud Orchestrator provide the
following benefits:
򐂰 Coordinate services and tasks such as compute and storage provisioning, configuration of
network devices, integration with service request and change management systems, and
integration with data center tools and processes.
򐂰 Allow cloud administrators to expose cloud services in a simpler self-service portal for
user consumption.
򐂰 Control image sprawl, manage image drift, and reduce security vulnerability through
analytics, image versioning, and federated image library features.
򐂰 Include monitoring and capacity analytics capabilities to help consolidate and balance
workloads.
򐂰 Facilitate measuring the cost of cloud services with metering and charge-back capabilities.
2.5.4 VMware
VMware provides servers virtualization on Intel based architecture. The core components of
the VMware solution are as follows:
򐂰
򐂰
򐂰
򐂰
VMware ESX and ESXi based hypervisor
VMware vSphere vCenter for providing management capabilities
vSphere vMotion to combat planned downtime
Vmware vCenter Site Recovery Manager to automate end-to-end recovery processes for
virtual applications
Figure 2-12 on page 47 shows the vSphere suite in a more comprehensive way. vSphere is a
product suite similar to Microsoft Office suite, which contains Microsoft Office Word, Excel,
Access, PowerPoint, and so on. VMware vSphere suite includes an ESXi hypervisor, vCenter,
and vSphere client. ESXi is a hypervisor installed on a physical machine. The vSphere client
is installed on the VMware administrator’s notebook or desktop computer and is used to
access the ESXi server to install and manage virtual machines on ESXi server. The vCenter
46
IBM SmartCloud Virtual Storage Center
server is installed as virtual machine on top of ESXi server. vCenter server is a vSphere
component which is mostly used in large environment where there are many ESXi servers
and several virtual machines. The vCenter server can also be accessed by vSphere client for
management purpose. So, vSphere client is used to access ESXi server directly in a small
environment; in larger environment, vSphere client is used again to access the vCenter
server, which ultimately manages the ESXi server.
.
Figure 2-12 VMware vSphere Suite overview
VMware servers hypervisor
VMware ESX and VMware ESXi are hypervisors that allow you to abstract processor,
memory, storage, and networking resources into multiple virtual machines (VMs) that can run
unmodified operating systems and applications. VMware ESX and VMware ESXi are
designed to reduce server sprawl by running applications on virtual machines that consist of
fewer physical servers. VMware ESX and VMware ESXi hosts can be organized into clusters.
This configuration allows ESX to provide flexibility in terms of what virtual machines are
running on what physical infrastructure.
VMware vCenter
vCenter is the management software suite that is used to manage the virtual machines inside
an ESX or ESXi host. When you allocate resources such as memory, storage, networking, or
processors to a virtual machine, a vCenter server manages how these resources are
allocated and maintained. vCenter can manage single ESX or ESXi hosts and clusters of
hosts. VMware vcenter has several features that allow for mobility of VMs between ESX hosts
and storage. These features can add to the availability of the VMs running in a cluster.
Chapter 2. Virtual Storage Center (VSC)
47
VMware vMotion
vMotion is a technology that is designed to combat planned downtime. vMotion is used to
move VMs between host and data stores to allow scheduled maintenance procedures to
proceed without affecting VM availability or performance. It is included in the Enterprise and
Enterprise Plus versions of VMware vSphere. It is shown in Figure 2-13.
Figure 2-13 VMware vMotion
VMware Host vMotion
Host vMotion eliminates the need to schedule application downtime for planned server
maintenance. It does so through live migration of virtual machines across servers with no
disruption to users or loss of service.
This process is managed from a vCenter server, which maintains client or application access
to a VM while it is moving between physical servers. In a SAN Volume Controller stretched
cluster solution, this feature is useful for moving VMs between two failure domains. You might
need to move VMs to load-balance across failure domains or because a failure domain needs
an outage for maintenance.
VMware Storage vMotion
Storage vMotion eliminates the need to schedule application downtime because of planned
storage maintenance or during storage migrations. It does so by enabling live migration of
virtual machine disks (VMDK) with no disruption to users or loss of service. The vCenter
server manages the copy of data from one data store to another. With vStorage APIs for Array
Integration (VAAI), this process can be off loaded to the storage system, saving resources on
both the vCenter host and data network.
48
IBM SmartCloud Virtual Storage Center
Figure 2-14 illustrates use of VMware Storage vMotion in a SAN Volume Controller stretched
cluster solution.
Figure 2-14 VMware Storage vMotion
In a SAN Volume Controller stretched cluster solution, this feature is useful for moving a
virtual machine’s VMDK file between two systems. You might move this file to ensure that it is
on the same failure domain as the VM, or to migrate off a storage device that is becoming
obsolete or is undergoing maintenance, as shown in Figure 2-14.
For more information about Storage vMotion, see the following web page:
http://www.vmware.com/files/pdf/VMware-Storage-VMotion-DS-EN.pdf
VMware vCenter Site Recovery Manager
Site Recovery Manager integrates with VMware vCenter server, and underlying storage
replication products, to automate end-to-end recovery processes for virtual applications. It
provides a simple interface for setting up recovery plans that are coordinated across all
infrastructure layers. Recovery plans can be tested nondisruptively as frequently as required
to ensure that the plan will meet availability objectives. At the time of a failure domain failover
or migration, Site Recovery Manager automates both the fail over and fail back processes. It
ensures fast and highly predictable RPOs and RTOs.
For more information about vCenter Site Recovery Manager, see the following web page:
http://www.vmware.com/products/site-recovery-manager/overview.html
Chapter 2. Virtual Storage Center (VSC)
49
VMware Distributed Resource Scheduler
Distributed Resource Scheduler (DRS) dynamically balances computing capacity across a
collection of hardware resources that are aggregated into logical resource pools. It
continuously monitors utilization across resource pools and intelligently allocates available
resources among the VMs that are based on predefined rules that reflect business needs and
changing priorities. When a VM experiences an increased load, VMware DRS automatically
allocates more resources by redistributing VMs among the physical servers in the resource
pool.
VMware DRS migrates and allocates resources by using a set of user-defined rules and
policies. These rules and policies can be used to prioritize critical or high performing VMs,
ensure that particular VMs never run on the same storage or host, or save on power and
cooling costs by powering off ESX servers that are not currently needed.
For more information about Distributed Resource Manager, see the following web page:
http://www.vmware.com/pdf/vmware_drs_wp.pdf
VSC and VMware integration
VSC and VMware are integrated by using Tivoli Storage Productivity Center plug-ins, as
shown in Figure 2-15.
Figure 2-15 VSC and VMware integration topology
50
IBM SmartCloud Virtual Storage Center
VMware vCenter web client extension provides the following features to VMware
administrators:
򐂰 Visualization of connections:
– End-to-end from storage volume to VM
– Storage details like pools, volumes, attributes
– Performance charts
򐂰 Automated policy-based storage provisioning, based on storage services catalog:
– Block volumes
– File shares
– Service class characteristics (storage tier, data protection, compression, encryption,
and so on)
򐂰 vSphere API for Storage Awareness (VASA):
– Alerts for performance, errors and capacity thresholds
– Availability of volumes, pools, storage systems, paths
– Tivoli Storage Productivity Center service classes can be used as VASA capabilities
򐂰 Additional storage reports are available in vCenter admin GUI:
– Fabric connectivity
– Storage performance
– Storage mappings
2.6 IBM Virtual Storage Center offerings
IBM SmartCloud Virtual Storage Center V5.2 has the following offerings:
򐂰 IBM SmartCloud Virtual Storage Center V5.2
򐂰 IBM SmartCloud Virtual Storage Center Entry V5.2
򐂰 IBM SmartCloud Virtual Storage Center for Storwize Family V5.2
The IBM SmartCloud VSC V5.2 license is an offering to be used with the System Storage
SAN Volume Controller and is a software entitlement to run both the external virtualization,
FlashCopy and remote copy services features. The only feature of the SAN Volume Controller
that is not included in the IBM SmartCloud VSC V5.2 license entitlement is the Real-Time
Compression option, which can be added as a separately priced license. This license does
not include the hardware nodes that are required for a complete SAN Volume Controller
implementation.
IBM SmartCloud VSC Entry V5.2 provides external virtualization, FlashCopy, and remote
copy services software entitlement in smaller SAN Volume Controller configurations. Also, for
deployment in midrange environments, a Storwize V5000 or V7000 can be used as the
virtualization engine in a VSC configuration, and in this case the offering to be used is the
IBM SmartCloud Virtual Storage Center for Storwize Family V5.2.
The versions of code that are available through IBM SmartCloud VSC 5.2 for download for the
System Storage SAN Volume Controller and the Tivoli Storage FlashCopy Manager are
exactly the same as the versions available for download if these products were downloaded
independently of IBM SmartCloud VSC 5.2. In the case of Tivoli Storage Productivity Center,
the code is the same as the independent product, but the VSC license enables the Storage
Advanced Engine functions to be used.
Chapter 2. Virtual Storage Center (VSC)
51
2.6.1 License model overview
IBM Virtual Storage Center can help customers to easily migrate their storage to a virtual
environment and manage storage efficiently. IBM VSC licensing charges are based on the
entire managed capacity. This contrasts with SAN Volume Controller, where FlashCopy and
Metro Mirror or Global Mirror can be licensed on virtual capacity for those functions only. The
managed capacity model avoids over-provisioning, which can become expensive with SAN
Volume Controller. Table 2-2 compares the current IBM VSC and IBM Tivoli Storage
Productivity Center licensing options and features. The sections after the tables have more
details about each of the IBM Virtual Storage Center licenses.
Table 2-2 Current 5.x VSC and Tivoli Storage Productivity Center licensing by offerings
Product name
Licensing usage
Tivoli Storage
Productivity
Center license
VSC
Per terabyte
(greater than 250 TB or
greater than two I/O groups)
For example, with the VSC
license, you can have 100 TB
and grow to 300 TB. This is
not possible with VSC Entry,
which is limited to less than
250 TB)
VSC Entry
FlashCopy
Manager
license
SAN
Volume
Controller
licensea
Tivoli Storage
Productivity Center
Advanced


Per terabyte
(less than 250 TB and less
than two I/O groups)
Tivoli Storage
Productivity Center
Advanced


VSC for Storwize
Family
Per enclosure
Tivoli Storage
Productivity Center
Advanced

Tivoli Storage
Productivity Center
Per terabyte
Tivoli Storage
Productivity Center
Tivoli Storage
Productivity Center
Select
Per enclosure
Tivoli Storage
Productivity Center
Select
Storwize
licenseb

a. SAN Volume Controller License includes Base, and FlashCopy and Remote Copy (Metro Mirror and Global
Mirror) license.
b. Storwize license included in VSC for Storwize Family is for external virtualization only. The base virtualization
license must be configured for each Storwize enclosure as usual.
52
IBM SmartCloud Virtual Storage Center
For customers who might be migrating from an older IBM Tivoli Storage Productivity Center to
a current IBM Virtual Storage Center or IBM Tivoli Storage Productivity Center license,
Table 2-3 compares functional features of the various licenses.
Table 2-3 Comparison of licenses by feature
Tivoli Storage Productivity Center 4.2.x
Feature
Discovery, topology,
monitoring, capacity
management,
alerting, basic
capacity reporting
Disk performance
reporting
Basic
Disk
Data





Repa
Stdb
Tivoli Storage Productivity Center (TPC)
or VSC 5.x
TPC and
TPC Select
VSC and
VSC Entry
VSC for
Storwize Family








2-site replication
management




3-site replication
management




Advanced NAS
functions




File System & DB
scans, reports,
quotas, script
integration






Fabric performance
reporting

Storage Optimizer



SAN Storage Planner
with policy based
provisioning



VMWare Hypervisor
reports



Storage tier reports





Cloud API (Tivoli
Storage Productivity
Center internal)


SAN Volume
Controller: Base,
FlashCopy and
Remote Copy (Metro
Mirror and Global
Mirror) license

Web GUI and Cognos



Chapter 2. Virtual Storage Center (VSC)
53
Tivoli Storage Productivity Center 4.2.x
Feature
Basic
Disk
Data
Repa
Stdb
Tivoli Storage Productivity Center (TPC)
or VSC 5.x
TPC and
TPC Select
VSC and
VSC Entry
VSC for
Storwize Family

Storwize: External
Virtualization,
FlashCopy and
Remote Copy (Metro
Mirror and Global
Mirror) license
FlashCopy Manager


a. Replication
b. Standard
Note: VSC for the Storwize Family (5608-ACL) can also be used for the FlashSystem V840
internal storage enclosures. You must use VSC Standard (5608-AE1) for V840 externally
virtualized storage. If VSC is not required, you can continue to license FlashSystem V840
external virtualization (5639-FV7) per enclosure for externally virtualized storage.
2.6.2 VSC License: Enterprise deployment
Do you have a medium to large storage configuration (250 TB or greater), possibly with a
variety of storage systems to manage? And, is high performance on that virtualized storage
mandated? Consider using the Standard VSC license.
License: Standard VSC
This license offers these features:
򐂰 Per terabyte price metric with tiered capacity pricing
򐂰 No restrictions on terabyte and number of SAN Volume Controller I/O groups
򐂰 SAN Volume Controller is the storage engine of choice
򐂰 Includes all features of standard VSC license (virtualization, FlashCopy, mirroring,
Advanced Analytics)
Choose this VSC license when you are using an SAN Volume Controller as your storage
virtualization engine, your managed capacity is 250 TB (or greater), or you are using more
than two SAN Volume Controller I/O groups.
54
IBM SmartCloud Virtual Storage Center
2.6.3 VSC Entry License
Are you managing a smaller storage configuration (less than 250 TB), possibly with a variety
of storage systems to manage. And, is high performance on that virtualized storage
mandated? Consider using the VSC Entry license.
License: VSC Entry
This license offers these features:
򐂰 Per TB price metric
򐂰 Restricted to 250 TB1 and two SAN Volume Controller I/O groups
򐂰 SAN Volume Controller is the storage engine of choice
򐂰 Includes all features of standard VSC license (virtualization, FlashCopy, mirroring,
Advanced Analytics)
Choose the VSC Entry license when you are using a SAN Volume Controller as your storage
virtualization engine, your managed capacity is less than 250 TB, and you are using two or
fewer SAN Volume Controller I/O groups.
2.6.4 VSC for Storwize Family license
Are you managing a small to medium storage configuration (100 TB - 1 PB) where the
storage virtualization investment is largely with Storwize V7000 or Storwize V5000, which
might manage some variety of storage systems under them? Consider using the VSC for
Storwize Family license.
License: VSC for Storwize Family
This license offers these features:
򐂰 Restricted to deployment on Storwize V7000 and V5000 hardware
򐂰 Per Enclosure price metric
򐂰 No restrictions on the number of enclosures
򐂰 Includes all features of VSC (external virtualization, Mirroring, Advanced Analytics);
򐂰 License does not include base software license for Storwize enclosures
Choose the VSC for Storwize Family license if you are using a Storwize V7000 or Storwize
1
Use Real-Time Compression to extend capacity management beyond 250 TB.
Chapter 2. Virtual Storage Center (VSC)
55
56
IBM SmartCloud Virtual Storage Center
3
Chapter 3.
Storage management
This chapter describes the storage management component of IBM SmartCloud Virtual
Storage Center (VSC), also known as IBM Tivoli Storage Productivity Center. We discuss
how the storage management component is integrated in Virtual Storage center.
As an end-to-end storage management solution included in VSC, Tivoli Storage Productivity
Center provides a single point of control that helps administrators to manage every aspect of
the storage infrastructure, between the hosts, through the fabric, and down to the physical
disks across multi-site storage environments.
In this chapter, the following Tivoli Storage Productivity Center functions are described:
򐂰 End-to-end management of storage infrastructures
򐂰 Management of storage replication
򐂰 Management and administration of SAN-attached storage
򐂰 Optimization, provisioning, and transformation of SAN-attached storage
򐂰 Performance management of SAN-attached devices
򐂰 Monitoring, management, and control of SAN fabric components
򐂰 Capacity utilization and availability management of file systems and databases
We also discuss scenarios and installation considerations, with the primary focus on installing
a new Tivoli Storage Productivity Center and upgrading existing Tivoli Storage Productivity
Center installations to Tivoli Storage Productivity Center VSC edition.
© Copyright IBM Corp. 2015. All rights reserved.
57
3.1 Overview
IBM SmartCloud Virtual Storage Center enables IT storage administrators to easily migrate
an existing storage environment to an agile, cloud-based, software-defined storage
environment, and manage it effectively. By using Tivoli Storage Productivity Center as a
storage management solution, storage administrators can manage complete SAN and
storage infrastructures, and use advanced storage analytics to optimize, provision, and
transform storage efficiently.
Note: Advanced storage analytics functions (cloud configuration, optimization, provisioning
and transformation) are licensed only in IBM Tivoli Storage Productivity Center - Virtual
Storage Center Edition, apart from the other storage management features, which are also
available in standard IBM Tivoli Storage Productivity Center.
Figure 3-1 shows Tivoli Storage Productivity Center environment with its interfaces and
examples of monitored objects. The Tivoli Storage Productivity Center server represented at
the center of the drawing shows a single server installation on which database, reporting, and
the Tivoli Storage Productivity Center server components are installed.
TPC Web GUI
TPC Reporting
TPC-R Server
Active Directory
LDAP
VMware
Tivoli Storage Productivity Center Server
Cloud and IT Optimization
products
•
•
•
•
Other IBM Products
•
SmartCloud Storage Access
Web File
VM4 VM5
v840
FlashSystem
OpenStack Cinder Driver
VMWare VASA provider
Proxy
CIM-Agent
IBM Storage
Servers using
Storage Resource Agents
Generic
Block Storage
SAN Fabric
Figure 3-1 Example of Tivoli Storage Productivity Center configuration and components
58
VM3
SVC/Storwize
Interfaces
•
•
Mail
VM1 VM2
Cloud Orchestrator
SmartCloud Cost Manager
Tivoli Application Dependency
Discovery Manager
SmartCloud Monitoring
IBM SmartCloud Virtual Storage Center
Agentless
Servers
Simplified administration
Tivoli Storage Productivity Center provides simplified administration through the web-based
GUI, which enables consistency and control across complex storage infrastructures. It
simplifies many common tasks and enables users to drill down to element managers for
advanced tasks. The web GUI enables even non-experts to perform storage provisioning and
performance management tasks. Figure 3-2 shows the fully integrated, web-based user
interface.
Figure 3-2 Tivoli Storage Productivity Center web GUI
Storage management
Tivoli Storage Productivity Center provides device-level, integrated storage infrastructure
management capabilities for managing both IBM and non IBM storage systems.
Heterogeneous storage support is offered through the Storage Networking Industry
Association (SNIA) Storage Management Initiative Specification (SMI-S). Management
capabilities include operational control and provisioning of heterogeneous storage platforms,
including storage systems from EMC, NetApp, Hitachi, Oracle Solaris, and HP. In addition,
Tivoli Storage Productivity Center includes advanced operational control and performance
reporting for IBM storage systems that include DS8000, FlashSystem V840, Storwize family
systems, SAN Volume Controller XIV, and IBM GPFS™.
SAN management
Tivoli Storage Productivity Center SAN fabric management capabilities include automatic
resource and topology discovery, monitoring and alerts, zone control, and SAN
error-prediction capabilities. Adding to this Tivoli Storage Productivity Center also enables
performance management at the port and switch levels for SAN platforms from Brocade
Communications and Cisco Systems. Figure 3-3 on page 60 shows an example of
performance report at the port level.
Chapter 3. Storage management
59
Figure 3-3 Switch port performance
Servers capacity utilization
By using a Storage Resource Agent (SRA), Tivoli Storage Productivity Center provides
capacity utilization information including file and file system information about servers
(Figure 3-4). Tivoli Storage Productivity Center can also gather information about database
managers installed on the server, and NAS device information.
Figure 3-4 Server filesystem information
Tivoli Storage Productivity Center can also monitor servers without using the SRA agent.
Agentless server gives you the chance to model a host server, either a physical server or
virtual machine, in Tivoli Storage Productivity Center without deploying an SRA. This is useful
in those situations where you cannot or do not want to deploy an SRA (either for security
restrictions in providing administration credentials or simply to avoid loading a production
server with agent code). In Tivoli Storage Productivity Center, you can add servers without
deploying Storage Resource agents and still view the overall connectivity and capacity of
those servers.
60
IBM SmartCloud Virtual Storage Center
VMware integration
Tivoli Storage Productivity Center includes a VMware vCenter plug-in that can simplify
storage administration for VMware administrators. By using this plug-in, users and
administrators can provision storage directly from the VMware vSphere client GUI. Tivoli
Storage Productivity Center also provides details of storage capabilities to vSphere through
VMware APIs for Storage Awareness (VASA), which in turn can be used to define VMware
storage profiles. vSphere reports are enhanced with Tivoli Storage Productivity Center data
through VASA and vSphere storage report extensions that include the following information:
򐂰 Fabric details, including zones, switch and fabric
򐂰 End-to-end mapping of VMware storage for each virtual disk
򐂰 Storage system performance metrics on the total I/O rate and response times
Performance management
Tivoli Storage Productivity Center integrates built-in, context-sensitive performance
management that is accessible to any administrator. It collects information about the
performance of storage systems and switches. The information includes key performance
metrics and notifications of threshold violations that can help administrators to measure,
identify, and troubleshoot performance issues and bottlenecks in storage environment.
Performance management is server-centric so administrators can more easily visualize
storage performance problems from the user perspective. Administrators can quickly
generate performance graphs from several perspectives and align them with a simple mouse
click. Performance scenarios can be cloned, so administrators can review past performance
scenarios to better understand how problems first developed. Figure 3-5 shows an example
of a performance monitor.
Figure 3-5 Storage pools performance monitor
Replication management
Tivoli Storage Productivity Center replication management capabilities help simplify the data
replication process from end to end. It can automate complex replication tasks without scripts,
and can scale to support hundreds of replication sessions across thousands of volumes.
Chapter 3. Storage management
61
Tivoli Storage Productivity Center supports the following replication session types:
򐂰
򐂰
򐂰
򐂰
򐂰
򐂰
򐂰
IBM FlashCopy
IBM XIV Storage System snapshots
IBM Basic HyperSwap®
Metro Mirror
Global Mirror
Metro/Global Mirror
Global Copy
For replication management, Tivoli Storage Productivity Center can use IBM DS8000, XIV,
IBM System Storage SAN Volume Controller, IBM Storwize V7000, IBM Storwize V7000
Unified, and IBM Storwize V3700. Figure 3-6 shows Tivoli Storage Productivity Center
replication management GUI.
Figure 3-6 Tivoli Storage Productivity Center Replication management GUI
Tivoli Storage Productivity Center uses performance metrics and advanced analytics to
perform storage optimization which includes storage pool balancing, storage tier analysis,
and volume transformation. Storage optimization task optimizes the resources in your storage
environment by redistributing volumes across each tier, and by moving volumes to higher or
lower tiers. It also distributes the workload of volumes across pools on the same tier.
For more details about optimization, see Chapter 7, “Storage optimization” on page 171.
Tivoli Storage Productivity Center and SAN Volume Controller
integration
The foundation of SmartCloud Virtual Storage Center is the integration between the Tivoli
Storage Productivity Center and the SAN Volume Controller. Although the SAN Volume
Controller has a capable web-based management interface, some limitations exist in terms of
the reporting and visualization of the environment because the scope is limited to a single
SAN Volume Controller cluster.
Tivoli Storage Productivity Center is designed as an enterprise class storage management
product that reports on all the SAN Volume Controllers. It also reports on the back-end
storage being virtualized, the SAN infrastructure providing connectivity to the SAN Volume
Controller and the servers that are connected to the virtualized storage. Tivoli Storage
Productivity Center has been able to manage and report on SAN Volume Controller since
Tivoli Storage Productivity Center V3.1 and every release of Tivoli Storage Productivity
62
IBM SmartCloud Virtual Storage Center
Center since that initial release has incrementally added to the management and reporting
capabilities of the SAN Volume Controller. The result is that Tivoli Storage Productivity Center
now reports on more than one hundred distinct performance metrics related to the SAN
Volume Controller. Additional reports can help customers handle the reporting needs
regarding capacity and utilization of the storage space being virtualized, including
thin-provisioned volumes. In addition, Tivoli Storage Productivity Center can produce reports
based on a collection interval of hourly or daily views. The time span for the report is also
under complete control for historical trend analysis.
3.2 Scenarios and installation considerations
IBM SmartCloud Virtual Storage Center (VSC) bundles Tivoli Storage Productivity Center
with IBM System Storage SAN Volume Controller and Tivoli Storage FlashCopy Manager.
Installation of each component is independent to each other, but all the components, when
integrated together, provide a new agile, cloud-based, software-defined storage environment.
Depending on your environment and existing storage infrastructure, the following scenarios
are possible ways to install VSC components:
򐂰 Install VSC in a non IBM storage environment
In this scenario, VSC transforms your existing storage environment into a completely
virtualized storage environment by using SAN Volume Controller virtualization technology,
where Tivoli Storage Productivity Center will provide comprehensive end-to-end
management with advanced analytics. By using SAN Volume Controller advanced copy
services, FlashCopy Manager can provide fast, application-aware backups and restores.
򐂰 Install VSC in a mixed IBM and non IBM storage environment
In this scenario, VSC transforms your existing IBM and non IBM storage environment into
one consolidated, completely virtualized, storage environment by using SAN Volume
Controller virtualization technology, where Tivoli Storage Productivity Center will provide
comprehensive end-to-end management with advanced analytics. By using SAN Volume
Controller advanced copy services, FlashCopy Manager can provide fast,
application-aware backups and restores.
򐂰 Install VSC in an IBM environment without virtualization and management
In this scenario, VSC transforms your existing IBM storage environment into a completely
virtualized storage environment by using SAN Volume Controller virtualization technology
where Tivoli Storage Productivity Center will provide comprehensive end-to-end
management with advanced analytics. By using SAN Volume Controller advanced copy
services, FlashCopy Manager can provide fast, application-aware backups and restores.
򐂰 Install VSC in an IBM environment with virtualization and management
In this scenario, VSC will improve your existing IBM storage environment. Tivoli Storage
Productivity Center will add new advanced analytics functions to existing Tivoli Storage
Productivity Center functions and it will enable your environment to optimize and provision
storage. By using SAN Volume Controller advanced copy services, FlashCopy Manager
can provide fast, application-aware backups and restores.
In this section, we focus only on installing the Tivoli Storage Productivity Center in new
environments and in existing environments with an already-installed standard version of Tivoli
Storage Productivity Center. Also some suggestions about installing Tivoli Storage
Productivity Center on two or three sites for disaster recovery solutions are described.
Details about design considerations for the Tivoli Storage Productivity Center environment
are in IBM Tivoli Storage Productivity Center: Beyond the Basics, SG24-8236.
Chapter 3. Storage management
63
3.2.1 Installing new Tivoli Storage Productivity Center
In scenarios where Tivoli Storage Productivity Center is not installed, you must do a fresh
installation. The installation procedure includes these steps:
1.
2.
3.
4.
Access the software code and documentation
Evaluate and prepare hardware configuration
Evaluate system configuration and install prerequisite software
Install Tivoli Storage Productivity Center
Access the software code and documentation
Tivoli Storage Productivity Center code is available only as a download for registered users. A
refresh pack code is available at the IBM Fix Central support website:
http://www.ibm.com/support/fixcentral/
Depending on whether you are a customer, an IBM Business Partner, or working for IBM, the
way to obtain the code differs. Typically, you use IBM Passport Advantage® or IBM Partner
World. For more information about downloading the installation images for Tivoli Storage
Productivity Center by using Passport Advantage Online, see the following website:
http://www.ibm.com/support/docview.wss?uid=swg24038425
Although several packages exist for download, the three major parts are as follows:
򐂰 Tivoli Storage Productivity Center
򐂰 IBM DB2 Enterprise Server Edition
򐂰 Reporting:
– Tivoli Common Reporting/Cognos
– IBM Jazz™ for Service Management
– Webshere Application Sever
Optional parts which are part of the Tivoli Storage Productivity Center package but are not
required to install Tivoli Storage Productivity Center, might be required for added functionality
and can be configured later. The optional parts are as follows:
򐂰 Storage Resource Agent (SRA)
򐂰 IBM Tivoli monitoring
򐂰 Tivoli Storage Productivity Center Monitoring Agent for IBM Tivoli Monitoring
򐂰 IBM System Discovery Engine
After you download the code, extract all the compressed files before you begin the
installation. The suggestion is to extract Tivoli Storage Productivity Center images into the
same single temporary directory; all other parts can be extracted into separate temporary
directories for each part or group of parts (DB2, reporting, SRA, IBM Tivoli monitoring, IBM
System Discovery Engine).
64
IBM SmartCloud Virtual Storage Center
Requirements for Windows:
򐂰 On the Windows platform, if the installation images are downloaded to a directory that
has spaces or unusual special characters in the name, Tivoli Storage Productivity
Center does not install correctly. Also, the file path to the temporary directory cannot
exceed 90 characters in length.
򐂰 The requirement is for the installation images to be local on the server on which you will
install Tivoli Storage Productivity Center. If you mount a network drive with the Tivoli
Storage Productivity Center installation images and try to run the setup from there, it
will not work.
Table 3-1 lists all required Tivoli Storage Productivity Center 5.2.4 downloadable parts for
Windows installation. For optional monitoring parts and also AIX and Linux installation parts,
see the following web page:
http://www.ibm.com/support/docview.wss?uid=swg24038425
Table 3-1 Tivoli Storage Productivity Center 5.2.4 Windows packages
Part number
File name
Description
CN27PML
VSC_V5.2.4_WINDOWS_Part1.zip
Tivoli Storage Productivity Center Advanced
V5.2.4 Windows (1 of 2)
CN27DML
TPC_V5.2.4_WINDOWS_Part2.zip
Tivoli Storage Productivity Center V5.2.4
Windows (2 of 2)
CIZJ2ML
TPC_DB2_10.1.fp3a_WINDOWS.exe
DB2 ESE 10.1 Fix Pack 3a Windows AMD64
and Intel EM64T (X64)
CIXA4ML
ITCR_3.1.0.1_FOR_WINS.zip
Jazz for Service Management 1.1.0.3 for
Windows ML (Launchpad, PRS, Jazz
Repository, TDI)
CIN3KML
ITCR_3.1.0.1_FOR_CFM_WINS.zip
IBM Tivoli Common Reporting 3.1.0.1 for
Windows Multilingual
CIXA8ML
ITCR_3.1.0.2_FOR_WINS.zip
IBM Tivoli Common Reporting 3.1.0.2 for
Windows Multilingual
CIFS5ML
WAS_V8.5.0.1_FOR_JAZZSM_WIN_M
L.zip
IBM WebSphere® Application Server V8.5.0.1
for Jazz for Service Management for Windows
ML
CN27HML
TPC_V5.2.4_STORE_RES_AGENT_FO
R_W.zip
Tivoli Storage Productivity Center V5.2.4
Storage Resource Agent for Windows
Chapter 3. Storage management
65
Evaluate and prepare hardware configuration
Hardware configuration requirements for Tivoli Storage Productivity Center installation
depends on the environment to be managed. Tivoli Storage Productivity Center server can
require a large amount of memory, disk space, network bandwidth, and processor resources,
which must all be planned in advance. In general we can make the following classification
whereby the requirements will be defined:
򐂰 Evaluation environments
This can include 10,000 volumes, 5 subsystems, 5 switches, 200 fabric ports, and 300
agentless servers.
򐂰 Medium production environments
This can include 50,000 volumes, 20 subsystems, 15 switches, 800 fabric ports, and 1000
agentless servers.
򐂰 Large environments
This can include 70,000 volumes, 30 subsystems, 20 switches, 1000 fabric ports, and
2000 agentless servers.
Tivoli Storage Productivity Center can be installed in a single server or a multiple-server
environment. A multiple-server environment might be suitable for large environments, where
one server is not sufficient to manage the Tivoli Storage Productivity Center components.
Other considerations to help you decide about a single or multiple server installation are
customer policies on server location and database placement.
The preferred practice is single server installation for several reasons:
򐂰 The complexity of the installation and future upgrades is far less in a single server
environment.
򐂰 The service.bat/sh troubleshooting tool runs on the server where Tivoli Storage
Productivity Center is installed. Therefore in a multiple-server environment troubleshooting
might not be as easy as if Tivoli Storage Productivity Center is running on a single server,
and the service.bat/sh tool can collect all of the logfiles.
򐂰 Backing up a Tivoli Storage Productivity Center environment is much simpler if you must
only back up one server.
The preferred practice for placement of the DB2 database repository is to place it on the
same server where you install the Tivoli Storage Productivity Center.
Details and current information about Tivoli Storage Productivity Center hardware
requirements, according to the classification, is at the following web page:
http://www.ibm.com/support/docview.wss?uid=swg27039550
Evaluate system configuration and install prerequisite software
Before installing and using Tivoli Storage Productivity Center, you must evaluate system
configuration and software prerequisites. This includes supported operating systems,
supported web browsers for Tivoli Storage Productivity Center web GUI, supported DB2
database versions, SRA server requirements, CIM agent requirements, supported hypervisor
versions, multipath support, and more.
Before you start the Tivoli Storage Productivity Center installation, you must install the DB2
database which is a requirement. If you do start the Tivoli Storage Productivity Center
installation program before you install DB2 database, the installer inform you that DB2
database is not installed and that you must install it. Jazz for Service Management and Tivoli
Monitoring can be installed during the Tivoli Storage Productivity Center installation if you
66
IBM SmartCloud Virtual Storage Center
immediately plan to use reporting or it can be installed at any time later, after you install Tivoli
Storage Productivity Center.
SRAs can be installed on servers separately after you successfully install Tivoli Storage
Productivity Center. You can install an SRA either by using the Tivoli Storage Productivity
Center web GUI, or locally on the server by using a command.
CIM agents are used to collect data from third-party storage systems, which does not have
native API for Tivoli Storage Productivity Center communication. For all IBM storage systems,
we collect data by using the native API. The CIM agents are not required to be installed
before Tivoli Storage Productivity Center, but they must be installed and configured before
you can collect the data through Tivoli Storage Productivity Center.
Detailed information about the software prerequisites are at the following web page:
http://www.ibm.com/support/docview.wss?uid=swg27039833
Install Tivoli Storage Productivity Center
After you evaluate and prepare the hardware and software configuration, you can start the
Tivoli Storage Productivity Center installation program from the directory where you extracted
Tivoli Storage Productivity Center images. The installation wizard guides you through the
installation steps. After you enter all the information in the wizard, the preinstallation summary
panel shows a summary before the installation starts. When the installation is complete, in
the Installation Complete panel, the links to connect to Tivoli Storage Productivity Center and
Tivoli Storage Productivity Center for Replication web user interfaces are shown.
Detailed steps and all information about installing Tivoli Storage Productivity Center are at the
following IBM Knowledge Center web page:
http://www.ibm.com/support/knowledgecenter/SSNE44_5.2.4/com.ibm.tpc_V524.doc/fqz0_
t_installing_main.html
The Tivoli Storage Productivity Center installation process is also described in IBM Tivoli
Storage Productivity Center V5.2 Release Guide, SG24-8204.
3.2.2 Upgrading existing Tivoli Storage Productivity Center installation to
Tivoli Storage Productivity Center VSC edition
In situations where you want to upgrade an existing Tivoli Storage Productivity Center
installation to Tivoli Storage Productivity Center Virtual Storage Center Edition and to use
advanced analytics functions (such as optimization and provisioning), you can use the
installation wizard or silent mode installation to upgrade to Virtual Storage Center Edition.
Licenses and upgrades
Understand the license and upgrade information.
Determining the license type
To determine the type of license you have (this applies to Tivoli Storage Productivity Center
version 5.2), find these files in TPC_installation_directory/properties/version directory:
򐂰 Tivoli_Storage_Productivity_Center.5.2.0.swtag: Indicates that you have a Tivoli
Storage Productivity Center license.
򐂰 Tivoli_Storage_Productivity_Center_Advanced.5.2.0.swtag: Indicates that you have
IBM SmartCloud Virtual Storage Center Storage Analytics Engine license. Previously, this
license was called the Tivoli Storage Productivity Center Advanced license.
Chapter 3. Storage management
67
򐂰 Tivoli_Storage_Productivity_Center_Select.5.2.0.swtag: Indicates that you have a
Tivoli Storage Productivity Center Select license.
Depending on the license you purchased, find one of the following license key files in the
license\key subdirectory, on disk1, part1 of the DVD, or electronic image:
򐂰 nodelock: Tivoli Storage Productivity Center license
򐂰 nodelock.SEL: Storage Productivity Center Select license
򐂰 nodelock.AE: IBM SmartCloud Virtual Storage Center Storage Analytics Engine license
Possible upgrades
Depending on your existing installation, the following upgrades are possible:
򐂰 Upgrade the existing Tivoli Storage Productivity Center 4.2.x and Tivoli Storage
Productivity Center 5.1.x installation to Tivoli Storage Productivity Center 5.2 VSC edition,
which includes the advanced analytics license.
򐂰 Upgrade the existing Tivoli Storage Productivity Center 5.2 license to Tivoli Storage
Productivity Center 5.2 VSC edition license, which enables the advanced analytics.
Considerations for upgrading existing Tivoli Storage Productivity Center 4.2.x and Tivoli
Storage Productivity Center 5.1.x installation to Tivoli Storage Productivity Center 5.2 VSC
edition are described in these sections.
򐂰 “Considerations for both 4.2.x and 5.1.x upgrades” on page 68
򐂰 “Considerations for only 4.2.x upgrades” on page 69
򐂰 “Considerations for only 5.1.x upgrades” on page 70
However, for more details, see IBM Tivoli Storage Productivity Center V5.2 Release Guide,
SG24-8204.
Considerations for both 4.2.x and 5.1.x upgrades
These are the considerations for upgrading your Tivoli Storage Productivity Center from a
previous version to V5.2.4:
򐂰 Windows 2003 and Windows 2008 32-bit environments are not supported by Tivoli
Storage Productivity Center V5.2.4. To upgrade previous Tivoli Storage Productivity
Center installations, the operating system must first be migrated to a Windows 64-bit
environment.
򐂰 Tivoli Storage Productivity Center V5.2.4 minimum memory requirement is 12 GB. Verify if
additional memory is required for your Tivoli Storage Productivity Center server.
򐂰 DB2 9.7 32-bit version is not supported in Tivoli Storage Productivity Center V5.2.x.
DB2 must first be upgraded to the 64-bit version. For details about the DB2 upgrade, see
IBM Tivoli Storage Productivity Center Installation Guide, SC27-4058.
򐂰 The roles that were previously defined in Tivoli Storage Productivity Center were
consolidated in two roles: administrator and monitor. Check your user roles and change
them accordingly.
򐂰 In Tivoli Storage Productivity Center V5.2.4, IBM Tivoli Integrated Portal is not used.
During the Tivoli Storage Productivity Center upgrade process, you can choose to
preserve the old Tivoli Integrated Portal instance, for example, if it is being shared with
another product. Otherwise, we suggest you uninstall Tivoli Integrated Portal in order to
preserve system resources.
68
IBM SmartCloud Virtual Storage Center
򐂰 The LDAP configuration settings are not migrated from Tivoli Integrated Portal to Jazz for
Service Management. You must manually configure the Jazz for Service Management
with LDAP repository by using the WebSphere Application Server Administrative Console.
For details about the Jazz for Service Management configuration with LDAP repository,
see IBM Tivoli Storage Productivity Center Installation Guide, SC27-4058.
Important: If your DS8000 or SAN Volume Controller storage system is configured to
use Single Sign On with Tivoli Storage Productivity Center, then the DS8000 or SAN
Volume Controller has the Tivoli Integrated Portal information stored in their
configuration. This must be switched back to local authentication during the upgrade of
Tivoli Storage Productivity Center, because Tivoli Integrated Portal will no longer be
used and must be uninstalled.
To be safe, consider reverting to operating system authentication for the Tivoli Storage
Productivity Center and for the duration of the upgrade. Also, see your documentation
for the values to re-enable LDAP and Active Directory at a later time.
򐂰 Tivoli Integrated Portal data is not migrated. You must manually migrate the Tivoli
Common Reporting reports and any custom Tivoli Common Reporting reports. Consider
this information:
– If you configured Tivoli Common Reporting Version 2.1.1 by using the internal content
store, go to the following web page for migration steps:
http://www.ibm.com/support/knowledgecenter/SSEKCU_1.1.0/com.ibm.psc.doc_1.1.
0/tcr_original/ttcr_upgrading.html
– If you configured Tivoli Common Reporting Version 2.1.1 by using an external content
store such as DB2 and Oracle, go to the following web page for migration steps:
http://www.ibm.com/support/knowledgecenter/SSEKCU_1.1.0/com.ibm.psc.doc_1.1.
0/tcr_original/ttcr_upgrading_external.html
The migration procedures migrate the existing default reports definitions; they are not
overwritten by the new definitions. Therefore reports with duplicated name are shown.
More information:
򐂰 For details about how to customize the export of reports from Tivoli Storage Productivity
Center V5.1, see IBM Tivoli Storage Productivity Center V5.1 Technical Guide,
SG24-8053.
򐂰 For details about exporting BIRT reports, see the Tivoli Common Reporter User Guide,
SC14-7613.
Considerations for only 4.2.x upgrades
These are additional considerations for upgrading your Tivoli Storage Productivity Center
specifically from 4.2.x to V5.2.4:
򐂰 Tivoli Storage Productivity Center V5 does not support CAS based Data or Fabric agents.
You must migrate these agents to Storage Resource Agents (SRA). This can be done
either before migration (suggested) or after migration. For more information, see IBM Tivoli
Storage Productivity Center Version 5.1 Installation and Configuration Guide, SC27-4058.
򐂰 The Tivoli Storage Productivity Center V5 database schema changed for performance and
configuration history functions. Depending upon on the amount of history data you have, it
might take from a few minutes to several hours for the installer to transform this data. The
installer will provide a time estimate. You can choose to drop the history data during the
installation.
Chapter 3. Storage management
69
Considerations for only 5.1.x upgrades
This is an additional consideration for upgrading your Tivoli Storage Productivity Center
specifically from 5.1.x to V5.2.4:
򐂰 Tivoli Common Reporting 3.1.0.1 uses DB2 as the external content store. Tivoli Common
Reporting 2.1.1 uses Derby as the external content store by default. However, some Tivoli
Storage Productivity Center users configured Tivoli Common Reporting 2.1.1 to use DB2
as the content store after installation. These users must follow a special procedure to
install Tivoli Common Reporting 3.1.0.1, reusing the Tivoli Common Reporting 2.1.1
external content store. The procedure is described in the IBM Knowledge Center:
http://www.ibm.com/support/knowledgecenter/SSEKCU_1.1.0.1/com.ibm.psc.doc_1.1.0
.1/tcr_original/ttcr_upgrading_external.html
Example: Upgrading 5.2 license to 5.2 VSC edition license
Upgrading the existing Tivoli Storage Productivity Center 5.2 license to Tivoli Storage
Productivity Center 5.2 VSC edition license is described in this example.
Figure 3-7 shows the existing Tivoli Storage Productivity Center installation, which has a
basic license and which does not support optimization and provisioning.
Figure 3-7 Tivoli Storage Productivity Center without advanced analytics functions
To start the license upgrade of your existing Tivoli Storage Productivity Center installation,
complete the following steps:
1. Log on to your Tivoli Storage Productivity Center system with the appropriate user
privileges.
2. From the Install directory (for example C:\Install\TPC) where you extracted the VSC
package, start the installation program by right-clicking Setup, and selecting Run as
administrator.
70
IBM SmartCloud Virtual Storage Center
3. On the installation wizard page, select a language and click OK. The language that you
select is used to install the license.
4. On the Choose Installation Location and Type page (Figure 3-8), select License upgrade
and click Next.
Figure 3-8 Installation type
Chapter 3. Storage management
71
5. Specify the location of the new license and click next (Figure 3-9).
Figure 3-9 Specify new licence
72
IBM SmartCloud Virtual Storage Center
6. On the Preinstallation summary panel, click Install (Figure 3-10).
Figure 3-10 Licence upgrade summary
Chapter 3. Storage management
73
7. The installation starts. When it completes successfully, as shown in the Installation
Completed page, click Done (Figure 3-11).
Figure 3-11 Licence upgrade
74
IBM SmartCloud Virtual Storage Center
8. Stop and restart the Tivoli Storage Productivity Center web-based GUI. New advanced
analytics functions are available and you can use optimization and provisioning functions
from the GUI (Figure 3-12 and Figure 3-13).
Figure 3-12 Tivoli Storage Productivity Center with advanced analytics functions
Figure 3-13 Cloud configuration option in Tivoli Storage Productivity Center
Chapter 3. Storage management
75
3.2.3 Use case: Replication management, installing Tivoli Storage
Productivity Center in DR environments
If you plan to implement Tivoli Storage Productivity Center in your disaster recovery (DR)
environment and use replication management (the Tivoli Storage Productivity Center for
Replication component), consider installing Tivoli Storage Productivity Center on the primary
and disaster recovery sites. By using the Tivoli Storage Productivity Center for Replication
component of Tivoli Storage Productivity Center you can manage replication and automate
complex replication tasks.
If you do not use the Tivoli Storage Productivity Center for Replication component, you might
still consider installing, as “active” Tivoli Storage Productivity Center on the primary site and
as “standby” Tivoli Storage Productivity Center server on the DR site. In such configurations
the “active” Tivoli Storage Productivity Center server will manage the whole environment
(primary site and DR site); the “standby” Tivoli Storage Productivity Center server will become
active only in case of a primary site disaster. This requires restoring the Tivoli Storage
Productivity Center database.
Note: IBM Tivoli Storage Productivity Center: Beyond the Basics, SG24-8236 explains
backup and restore options of the Tivoli Storage Productivity Center database that reside
on the Tivoli Storage Productivity Center server.
For DR solutions, high availability configuration of Tivoli Storage Productivity Center for
Replication is highly suggested because it provides high availability of replication
management. As a preferred practice we suggest that the active Tivoli Storage Productivity
Center for Replication server is always the primary site; the standby server is required at the
disaster site. If you have a three-site solution, the standby server is not required in the
intermediate site; in a three-site solution, such as Metro Mirror and Global Mirror, the standby
server is required in the third site. Tivoli Storage Productivity Center for Replication does not
support two standby servers. In case of disasters on the primary site, the standby server will
be in consistent disconnected status and it will be ready to take over. This action causes the
standby server to be active and you can continue to manage replication.
3.3 Advanced storage analytics overview
As a storage management component of IBM SmartCloud Virtual Storage Center, Tivoli
Storage Productivity Center is also designed to provide advanced storage analytics for
storage optimization and provisioning capabilities. Those capabilities are used in
cloud-based, software-defined, storage environments where the customer's workload
requirements must be met in real time.
Tivoli Storage Productivity Center uses real performance metrics and advanced analytics to
make recommendations to optimize storage pools and volumes by redistributing workloads
across storage environment. By using real performance metrics, it enables optimization
decisions to be made based on actual usage patterns, rather than on predictions.
To take advantage of the optimization and provisioning capabilities that are available in Tivoli
Storage Productivity Center, some configuration is required. This configuration is called cloud
configuration and it specifies storage tiers, service classes, and capacity pools that are used
to transform your environment to an efficient, self-optimized, and cloud-agile environment.
76
IBM SmartCloud Virtual Storage Center
The following sections introduce Tivoli Storage Productivity Center advanced storage
analytics tasks. Also see the “Cloud configuration and provisioning” and “Storage
optimization” topics in IBM Tivoli Storage Productivity Center V5.2 Release Guide,
SG24-8204.
3.3.1 Cloud configuration
With the Tivoli Storage Productivity Center cloud configuration function, you can organize
your storage environment by categorizing the resources in tiers and defining service classes
with specific qualities of service. This function is a requirement in optimizing your storage
environment and in setting up a storage cloud implementation. It is also highly preferred in a
traditional storage provisioning context.
To provision storage, you specify only the storage capacity and storage quality that is
required. By using the service class, volumes are created with the required characteristics.
To optimize storage performance, Tivoli Storage Productivity Center analyzes tiering and,
according to the recommendations that are generated, volumes are redistributed across each
tier and can also be moved to higher or lower tiers.
Cloud configuration involves the following steps:
1.
2.
3.
4.
Assign the tier to storage pools.
Refine the customization by assigning tags to storage resource.
Define capacity pools.
Define service class.
Tiering
Storage tiering is a step toward defining service classes and provisioning volumes that
require a certain tier level. It also allows you to optimize the placement of volumes by using
Tivoli Storage Productivity Center optimization tasks, which redistribute volumes within the
tier or move the volumes to higher or lower tiers.
Tivoli Storage Productivity Center uses 10 tiering levels of storage pools: tier 1 represents the
best performing tier, and tier 10 the least performing tier. You can use as many levels as you
want, based on the storage systems and disk technologies in your environment. In the
storage pools tiering process, you might consider the technologies that are not present in
your storage environment, but that might be added in the near future. In this way, you can
leave space among tier levels for assignment later so that you do not need to reconfigure tiers
and service classes when new technologies are added to your environment.
For example, you might categorize the following storage tiers:
򐂰
򐂰
򐂰
򐂰
򐂰
򐂰
򐂰
򐂰
򐂰
򐂰
Tier 1. Leave this level available for future uses.
Tier 2. Flash disk pools and solid-state drive (SSD) disk pools
Tier 3. Leave this level available for future uses.
Tier 4. Fibre Channel and SAS disk pools
Tier 5. Leave this level available for future uses.
Tier 6. Hybrid pool
Tier 7. Leave this level available for future uses.
Tier 8. NL-SAS disk pool
Tier 9. Leave this level available for future uses.
Tier 10. SATA disk pool
Chapter 3. Storage management
77
Figure 3-14 shows an example of storage tiering in Tivoli Storage Productivity Center where
five tiers are defined.
Figure 3-14 Storage tiers
Service classes
A service class is a set of properties that describe capabilities and characteristics of storage
resources. A service class typically describes a particular quality of service, and is used
during provisioning to describe storage requirements. For example, a block-storage service
class specifies properties such as a required RAID level, storage tier, volume mirroring,
multipathing policy, and whether storage resources must be able to encrypt or thin provision
volumes.
Service classes simplify provisioning requests by representing a level or type of storage
quality. When you request storage, only the required capacity and service class must be
specified. Before you can provision storage, you must create service classes that describe
the capabilities and characteristics of the storage that you want to be able to provision.
Service classes can later be modified or deleted as the needs of your installation change.
Tivoli Storage Productivity Center accounts for all attributes of the service class, specified in a
provisioning request, and also current storage resource utilization (space, performance, and
status) in order to identify the most appropriate resource for the new volume or share.
Service classes are a key point in mapping business requirements (capacity, accessibility,
performance, and availability) in infrastructure capabilities (media type, disk technologies,
RAID Levels, encryption, compression, and thin provisioning).
Two types of service classes are available:
򐂰 Block-storage service classes:
This type describes attributes and capabilities of block storage resources. When you
provision volumes, you specify the requirements by using a block-storage service class.
Based on the requirements of the service class, Tivoli Storage Productivity Center
identifies a storage pool for the volume. Tivoli Storage Productivity Center provides three
predefined block-storage service classes. Although the classes are predefined, you might
want to change them or adjust the tiers, RAID levels, and so on:
– Gold: defined for mission-critical applications (highest-performing storage).
– Silver: defined for applications in production (high-performing storage).
– Bronze: defined for non-mission-critical applications (standard storage).
򐂰 File-storage service classes:
– NormalIsolation: (normal isolation file storage) The file system from which the NAS
share is provisioned can contain other NAS shares. Shared storage is allowed.
78
IBM SmartCloud Virtual Storage Center
– EnhancedIsolation: (enhanced isolation file storage) The file system from which the
NAS share is provisioned cannot contain other NAS shares. Dedicated storage is
required.
Figure 3-15 figure shows an example of defined service classes in Tivoli Storage Productivity
Center.
Figure 3-15 Predefined Services Classes in Tivoli Storage Productivity Center
Notes:
򐂰 For the gold configuration, the Virtualization on/off service class setting is important. If
Virtualization is set to “on,” only IBM storage virtualizers are candidates for provisioning.
If Virtualization is set to “off,” no IBM storage virtualizer is a candidate for provisioning.
򐂰 The Storwize systems have a layer setting that determines whether a Storwize system
can virtualize external storage or replicate with another Storwize system. You can use
the chsystem command with either the -layer replication or -layer storage option.
Setting this option to layer=storage prevents the service classes from including pools
as candidate pools.
To determine which storage systems (block and file) are supported for provisioning, see the
following web page:
http://www.ibm.com/support/docview.wss?&uid=swg21386446
Capacity pools
Capacity pools are logical groups of storage resources that include these items:
򐂰
򐂰
򐂰
򐂰
Storage systems
Storage pools
File systems of file storage systems
IBM SONAS Network Shared Disks (NSDs)
Capacity pools can group storage resources in any way that serves the needs of your
environment or business. By configuring capacity pools, you can track the used and available
space for block and file storage on any set of storage resources.
You can use capacity pools to define a set of storage resources from which provisioning
requests must be satisfied. When you request storage by using the Provision Storage wizard
or the vSphere Web Client extension, you can specify a capacity pool, and the provisioning
request is restricted to the resources in this capacity pool.
Chapter 3. Storage management
79
You can also associate a service class with capacity pools. If a service class is associated
with capacity pools, provisioning requests for the service class must be specified and be
constrained to one of the associated capacity pools. Because you can also grant to
non-administrative users permission to provision storage by using the service class,
associating capacity pools with the service class restricts the users to a specific set of
resources.
Optional: The use of capacity pools is optional.
Figure 3-16 shows an example of defined capacity pools in the Tivoli Storage Productivity
Center web GUI.
Figure 3-16 Capacity Pools
External clients integration with Tivoli Storage Productivity Center in
Cloud configuration
Tivoli Storage Productivity Center Cloud configuration is based on the Storage Management
API for Clouds (SMAC) API, which was introduced as a new API in Tivoli Storage Productivity
Center V5.1. In Tivoli Storage Productivity Center V5.2 it was integrated in the web GUI.
The Storage Management API for Cloud (SMAC) API allows integration of external clients
with Tivoli Storage Productivity Center. In this configuration the external clients can use
Service Classes and Capacity Pools which are defined in the Tivoli Storage Productivity
Center cloud configuration. Storage Management API for Cloud (SMAC) API is used by the
following external clients:
򐂰
򐂰
򐂰
򐂰
IBM Cloud Orchestrator
IBM SmartCloud Storage Access
Openstack Cinder driver
Tivoli Storage Productivity Center vSphere plug-in
The external clients use provisioning capabilities with Storage Management API for Clouds
(SMAC) API which includes Service Classes and optionally Capacity Pools to ease the
provisioning of volumes.
The Storage Management API for Cloud (SMAC) REST API is used by the Tivoli Storage
Productivity Center GUI and VMWare Tivoli Storage Productivity Center plugin. The VMWare
Tivoli Storage Productivity Center plugin use Storage Management API for Clouds (SMAC)
REST API for reporting capabilities.
80
IBM SmartCloud Virtual Storage Center
Figure 3-17 shows how the external clients integrate with Tivoli Storage Productivity Center.
Figure 3-17 Tivoli Storage Productivity Center - Storage Management API for Clouds (SMAC)
3.3.2 Storage provisioning
With the Tivoli Storage Productivity Center Advanced Analytics function, you can provision
storage resources to servers, hypervisors, or a cluster. You can provision storage resources
such as storage volumes or NAS file shares to one or more servers, one or more hypervisors,
or one cluster.
To provision the storage resources, you are required to specify only the storage capacity and
storage quality that is represented by using the service class definition. Tivoli Storage
Productivity Center identifies the storage resources that satisfy the requirements of the
specified service class from a pool of available resources. By default, the pool of available
resources includes all block storage or file storage resources that are known to Tivoli Storage
Productivity Center. If you specify a capacity pool, new storage volumes and NAS file shares
are provisioned only from storage resources in that capacity pool.
From the storage resources that can provide the required service class, Tivoli Storage
Productivity Center identifies the best storage placement. The determination of the best
storage placement is based on storage system free space. For provisioning storage volumes,
performance data is also considered. Preference is first given to storage pools and systems
that already contain volumes for the selected server, hypervisor, or cluster. Preference is then
given to systems that have available performance data.
Chapter 3. Storage management
81
Figure 3-18 shows the provisioning wizard in Tivoli Storage Productivity Center GUI.
Figure 3-18 Tivoli Storage Productivity Center provisioning
Provisioning is described in more detail in Chapter 6, “Provisioning” on page 141.
3.3.3 Optimization
By using storage optimization function in Tivoli Storage Productivity Center you can optimize
resources in your storage environment. The optimization function is used to help improve
performance of your storage environment and to better utilize storage resources. The
optimization function in Tivoli Storage Productivity Center has three main tasks:
򐂰 Balancing the workload of volumes across storage pools
򐂰 Re-tiering volumes
򐂰 Transforming volumes
Figure 3-19 on page 83 shows the optimization function in Tivoli Storage Productivity Center
web GUI.
Note: Tivoli Storage Productivity Center uses the VDisk copy function to implement
optimization recommendations. This means that the optimization tasks are supported with
only IBM SAN Volume Controller, IBM Storwize V7000, and Storwize IBM V7000 Unified.
82
IBM SmartCloud Virtual Storage Center
Figure 3-19 Optimization function
Balance pools
Balancing the workload of volumes across storage pools is done with balance pools function
where the workload of volumes is balanced across pools on the same tier. Balance pools is
used to analyze the pool performance and according to recommendations done on pool
activity, volumes are moved from pools with high-activity values to pools with low-activity
values in the same tier. Balance pools help you to achieve a more balanced workload across
the pools within same tier.
Analyze tiering
The analyze tiering function in Tivoli Storage Productivity Center is for optimizing the
placement of volumes on storage tiers. The purpose of analyzing tiering is to tier volumes
based on the criteria that you set in tiering policies. For example, you can tier volumes based
on the volume workload or on file usage, or both. Depending on the conditions that are set in
the tiering policy, recommendations are generated that you can implement to ensure that the
volumes are placed on the tiers that best match the workload requirements.
The analyze tiering function can help improve storage performance by moving volumes with
heavy workloads to the tiers that meet the workload requirements of the volumes. It also can
help you to reduce storage costs by moving volumes with low workloads from higher and
more expensive tiers to lower and less expensive tiers.
Volume transformation
The volume transformation function in Tivoli Storage Productivity Center is used to convert
volumes in storage pools to fully allocated volumes, compressed volumes, or thin-provisioned
volumes. It can be also used to move volumes to other pools on the same storage system or
to pools that are enabled for Easy Tier on the same storage system.
Chapter 3. Storage management
83
The following volume transformation options can be used:
򐂰 Migrate pool
Migrate pool option can move the selected volume on the recommended pool among
those that you selected from the available pools.
򐂰 Compression
The compression option is used to convert your fully allocated volume to compressed, or
to transform the compressed volume to a fully allocated volume.
򐂰 Thin provisioning
Thin provisioning enables the storage to present the required capacity to the host while
allocating only the actual used capacity in terms of space on the physical storage media.
By using the thin provisioning option, you can convert fully allocated volumes to
thin-provisioned volumes or thin-provisioned volumes to fully allocated volumes.
򐂰 Easy Tier
Easy Tier is a performance function that automatically migrates or moves extents off a
volume to, or from, one MDisk storage tier to, or from, another MDisk storage tier. Easy
Tier monitors the host I/O activity and latency on the extents of all volumes with the Easy
Tier function enabled in a multitier storage pool over a 24-hour period (heatmap creation).
Next, it creates an extent migration plan based on this activity and then dynamically moves
high activity or hot extents to a higher disk tier within the storage pool. It also moves
extents whose activity has dropped or cooled from the high-tier MDisks back to a
lower-tiered MDisk.
3.4 SAN and storage management
Tivoli Storage Productivity Center is designed to provide SAN and storage infrastructure
management capabilities such as automated system discovery, provisioning, configuration,
performance monitoring, and replication for storage systems. By using data collection
capabilities within managed environment Tivoli Storage Productivity Center provides assets
and status information, availability monitoring, usage and trending information and also
server monitoring.
Tivoli Storage Productivity Center provides, in a simple way, device management for multiple
storage arrays and storage area network (SAN) fabric components from a single integrated
console. Both IBM and non IBM storage systems (EMC, NetApp, Hitachi, Oracle Solaris and
HP) are supported through the Storage Networking Industry Association (SNIA) Storage
Management Initiative Specification.
84
IBM SmartCloud Virtual Storage Center
Figure 3-20 on page 85 and Figure 3-21 on page 85 show an example of SAN and storage
management in Tivoli Storage Productivity Center web GUI.
Figure 3-20 Storage management
Figure 3-21 SAN management
Chapter 3. Storage management
85
3.5 Performance management
Tivoli Storage Productivity Center integrates built-in, context-sensitive performance
management that is easily accessible from a single integrated console. Performance
management is server-centric and storage performance problems can be more easily
visualized from the user perspective. Performance graphs can be produced from several
perspectives and they can be aligned to help you better understand performance problems.
Tivoli Storage Productivity Center helps monitor and manage performance and measure
service levels by storing received performance statistics into database tables for future use.
Policy-based automation enables event action based on different business policies. You can
set Tivoli Storage Productivity Center performance thresholds for devices, based on selected
performance metrics, generating alerts when those thresholds are exceeded. These
capabilities can help you simplify the complex management of multiple SAN attached storage
devices.
Tivoli Storage Productivity Center performance management has these capabilities:
򐂰 Proactive performance management from a single, integrated console for monitoring
storage and SAN devices
򐂰 Monitoring of various metrics, such as I/O rates, throughput, cache utilization, back-end
storage utilization and so on
򐂰 Measuring and tracking of service levels by storing historical performance statistics for
analysis and report generation
򐂰 Generation of timely alerts that enable event action by setting performance thresholds
based on different business policies
Figure 3-22 shows an example of storage system performance.
Figure 3-22 Storage system performance
86
IBM SmartCloud Virtual Storage Center
3.6 Advanced monitoring and alerting
Tivoli Storage Productivity Center provides advanced monitoring and alerting for various
conditions and violations of performance thresholds that you defined in your storage
environment. Alerts are triggered by the thresholds or conditions that are detected during
data collection and event processing.
The conditions that trigger alert notifications depend on the type of resource that you are
monitoring. Some triggering conditions, like performance thresholds, require you to enter
values for triggering alerts. In general, the following types of conditions can trigger alerts:
򐂰
򐂰
򐂰
򐂰
A performance threshold was violated
A data collection job did not complete
A change occurred in the storage infrastructure
A change occurred in storage replication
When an event occurs and triggers an alert, the alert is written to a log. You can also select
one or more other ways to be notified of the event. These alert notifications include SNMP
traps, IBM Tivoli Enterprise Console® events, login notifications, entries in Windows event log
or UNIX syslog, and email.
Tivoli Storage Productivity Center shows you all defined alerts in the web GUI (Figure 3-23).
Figure 3-23 Alerts
Chapter 3. Storage management
87
88
IBM SmartCloud Virtual Storage Center
4
Chapter 4.
Storage virtualization
This chapter provides information about the three options available for the storage
virtualization component of the IBM SmartCloud Virtual Storage Center:
򐂰 IBM System Storage SAN Volume Controller
򐂰 IBM Storwize V5000
򐂰 IBM Storwize V7000
A comparison between those three options is made so that the people involved in an IBM
SmartCloud Virtual Storage Center project can decide which one best fits their business
needs, based on criteria such as:
򐂰 Performance
򐂰 Scalability
򐂰 Particular capabilities
© Copyright IBM Corp. 2015. All rights reserved.
89
4.1 External storage virtualization and software-defined
storage
IT environments are evolving to a model where higher levels of efficiency, availability, and
flexibility are mandatory characteristics of the infrastructure. New workloads and business
needs make infrastructure virtualization a faster way to achieve those goals compared to
using islands of resources.
Server virtualization has been adopted for a long time to help companies make better use of
their assets. Capabilities like virtual machines being moved nondisruptively between physical
servers, single point of control and management, resource pooling, agility when deploying
new applications, and so on, make server virtualization the foundation for new service
delivery models, such as cloud computing or software defined environments.
However, server virtualization is only part of the story. A need also exists to virtualize storage,
so that the entire infrastructure can achieve the business goals. Additionally, storage
virtualization is the bridge between the storage hardware (or data plane) and the other
services management modules (or control plane) of the software-defined storage (SDS)
framework (for more details about SDS, see 1.2.2, “Software-defined storage overview” on
page 17).
In the following topics, the characteristics of storage virtualization using IBM SAN Volume
Controller, Storwize V5000, and Storwize V7000 are discussed, with a focus on their
integration with the other IBM SmartCloud Virtual Storage Center components.
4.1.1 Product overview and integration with IBM SmartCloud Virtual Storage
Center
As a core part of the IBM SmartCloud Virtual Storage Center, the virtualization component
plays several important roles:
򐂰 Provides flexibility for the storage infrastructure, allowing multi-vendors and pools that are
in both faster more expensive storage and slower less expensive storage to be managed
as one global resource.
򐂰 Enables a common set of advanced functions to be applied to the storage infrastructure,
even if the individual hardware does not support advanced functions, such as Thin
Provisioning, Easy Tier, Real-time Compression, among others.
򐂰 Integrates with the management component so the entire storage environment has a single
point of control for monitoring, provisioning, and automation
򐂰 Provides a single platform for the application-aware data protection component to
establish point-in-time copies that go beyond the scope of individual disk subsystems.
򐂰 Enables the advanced analytics component optimization recommendations to be
executed, moving volumes between different pools and tiers nondisruptively to optimize
the workload, consisting of heterogeneous vendors and technologies
The following products, with these external storage virtualization capabilities, are available
with IBM SmartCloud Virtual Storage Center:
򐂰 IBM SAN Volume Controller
򐂰 IBM Storwize V5000
򐂰 IBM Storwize V7000
90
IBM SmartCloud Virtual Storage Center
The products have a common Storwize software platform so, independently of the size of the
environment, the user experience is mostly the same. For more details about the software
within the Storwize family of products, see 4.3, “Storwize software stack” on page 93.
4.2 Benefits of storage virtualization
Virtualization is a foundational technology for both cloud and software-defined environments
because of the larger number of benefits it brings to IT infrastructure. The following topics
discuss the many ways storage virtualization helps organizations obtain the best benefits
from their investment.
4.2.1 Operational efficiency
With the virtualization of the physical resources, the utilization of storage capacity grows from
an average of 50% up to 90%1. This can be achieved by enabling online data migration for
load balancing and easily moving volumes to the most cost-effective tier. Storage
virtualization also enables the provisioning of capacity to application servers from any of the
storage pools, independently of where the physical capacity resides in the global repository.
Other technologies are also introduced by storage virtualization, which apply to all the
back-end devices, even if they do not natively support those functions:
򐂰
򐂰
򐂰
򐂰
򐂰
Thin provisioning, for better physical capacity utilization
Real-time Compression, to store more in the same space
Automated tiering (Easy Tier), to optimize the investment in flash technology
Single point of management using an award-winning, user-friendly interface
Flexibility to add any supported back-end storage that better matches price/performance
requirements
IBM SmartCloud Virtual Storage Center leverages those benefits and complements them with
analytics capabilities, which help make better use of all the storage tiers, both homogeneous
and heterogeneous (when Easy Tier is enabled). Alternatively, storage virtualization enables
the IBM SmartCloud Virtual Storage Center analytics to execute the optimization
recommendations transparently moving data between tiers or transforming volumes.
4.2.2 Improved return of investment
Storage virtualization with IBM SAN Volume Controller and the Storwize family has a
comprehensive support matrix for back-end devices, allowing customers not only to build new
environments with heterogeneous systems, but also to virtualize what they already have
installed in their data centers. When reusing existing devices, unused space in the file
systems can also be reclaimed.
1
According to IBM storage infrastructure optimization assessments, 2011-2012:
http://www-03.ibm.com/employment/execjobs/files/GMU_External_Presentation.pdf
Chapter 4. Storage virtualization
91
Support matrix information:
򐂰 The support matrixes for each version of the IBM SAN Volume Controller software are
at the support portal web page:
http://www.ibm.com/support/docview.wss?uid=ssg1S1003658
򐂰 For the Storwize V7000, the support matrixes are at the following web page:
http://www.ibm.com/support/docview.wss?uid=ssg1S1003741
Software licensing can be reduced by using storage virtualization, since it concentrates all the
functions in the same layer, and individual features in the storage devices do not need to be
licensed anymore. These features include remote mirroring, point-in-time copy, thin
provisioning, and other licenses charged on a one-time basis or on a monthly basis.
Multipathing drivers that are charged by some vendors can also be reduced or eliminated.
IBM external storage virtualization solutions use drivers that are embedded in the operating
systems almost all the time, and IBM provides a no-charge complement for some drivers.
4.2.3 Enhanced data availability
When virtualizing the storage environment, advanced copy services are centralized in the
virtualization layer, making it possible to protect volumes in a heterogeneous infrastructure,
with multiple vendors and technologies. Snapshots can be moved from one storage device to
another, and remote replication can be done by using different vendors in each data center.
Data can be protected more easily and its recovery is faster.
IBM SAN Volume Controller can also be implemented in a stretched cluster architecture,
which isolates the hosts from individual disk subsystem or SAN failures. This solution
distributes the hardware components between two data centers and delivers continued
access to a logical volume which becomes available to hosts on both sites simultaneously.
Stretched cluster solutions: For more information, see these Redbooks publications:
򐂰 IBM System Storage SAN Volume Controller and Storwize V7000 Replication Family
Services, SG24-7574
򐂰 Implementing the IBM System Storage SAN Volume Controller V7.4, SG24-7933
򐂰 IBM SAN and SVC Stretched Cluster and VMware Solution Implementation,
SG24-8072
򐂰 IBM SAN Volume Controller Stretched Cluster with PowerVM and PowerHA,
SG24-8142
򐂰 IBM SAN Volume Controller Enhanced Stretched Cluster with VMware, SG24-8211
With IBM SAN Volume Controller and Storwize family solutions, volumes can be migrated
transparently from one storage pool to another, even if their data resides in different disk
subsystems. This way, applications do not suffer down time when data migrations are done,
for example, to isolate a device for maintenance, to replace an old subsystem by a new one,
to aggregate flash technology to the infrastructure, or simply relocating data.
92
IBM SmartCloud Virtual Storage Center
4.2.4 Transparent data mobility
Besides the great contribution to data availability as described in 4.2.3, “Enhanced data
availability” on page 92, transparent volume migration also provides a way to avoid overtime
of the storage management staff. As it occurs with server hypervisors which can move virtual
machines between physical servers nondisruptively, volume migration can also be done
within the production shift, or at least be initiated just before leaving, because it runs with no
need of human intervention.
Another way to migrate data with no impact to availability is by using volume mirroring, which
adds a second copy to a volume, synchronizes the volumes, and then the original copy can
be removed when it is no longer needed. This function also enables the volume
transformation function of the IBM SmartCloud Virtual Storage Center.
4.3 Storwize software stack
SAN Volume Controller and the Storwize family of products combine software and hardware
into comprehensive, modular appliances that use symmetric virtualization in a cluster
architecture. The Storwize software, which is a common platform to SAN Volume Controller
and Storwize family members, performs the following functions for the host:
򐂰
򐂰
򐂰
򐂰
Creates a single pool of storage, including capacity from external subsystems.
Provides logical unit virtualization.
Manages logical volumes with multipathing capabilities.
Mirrors logical volumes.
Note: External virtualization is only supported by SAN Volume Controller, Storwize V7000
and Storwize V5000.
SAN Volume Controller and Storwize systems also provide the following functions:
򐂰 Large scalable cache management
򐂰 Transparent data migration
򐂰 Copy services, including point-in-time copy, synchronous and asynchronous remote copy,
and volume mirroring
򐂰 Space management with Easy Tier, thin-provisioned logical volumes and real-time
compression
All these functions are implemented in a common software stack, as shown in Figure 4-1 on
page 94. This architecture applies to Storwize software version 7.3 and later.
Chapter 4. Storage virtualization
93
Figure 4-1 Storwize V7.3 software stack
The main difference compared to previous versions of the software, as highlighted in the
diagram, is the cache re-architecture, where the upper cache and the lower cache now allow
for these benefits:
򐂰 More independent implementation of new functions affecting the front or the back-end
򐂰 Higher scalability of the system in the future
򐂰 Improved performance of previous functions such as FlashCopy, Thin Provisioning,
Real-time Compression, and Volume mirroring
4.4 SAN Volume Controller component overview
The SAN Volume Controller product provides block-level aggregation and volume
management for attached disk storage. In simpler terms, the SAN Volume Controller
manages several back-end storage controllers or locally attached disks, and maps the
physical storage within those controllers or disk arrays into logical disk images, or volumes,
that can be seen by application servers and workstations in the SAN.
The SAN is zoned so that the application servers cannot see the back-end physical storage,
which prevents any possible conflict between the SAN Volume Controller and the application
servers both trying to manage the back-end storage. In this section, we briefly explain basic
architecture components of SAN Volume Controller.
94
IBM SmartCloud Virtual Storage Center
Nodes
Each SAN Volume Controller hardware unit is called a node as it pertains to a cluster, or
system. The node provides the necessary resources to the virtualization of a set of volumes,
such as cache, processors, and I/O ports. It can optionally provide compression accelerator
cards to help Real-time Compression processing. There is also an option to connect
expansion drawers to the nodes to accommodate flash drives.
SAN Volume Controller nodes are deployed in pairs, or I/O groups, and multiple pairs make
up a clustered system, also referred to as simply system. A system can consist of between
one and four SAN Volume Controller node pairs.
One of the nodes within the system is known as the configuration node. The configuration
node manages the configuration activities for the system. If this node fails, the system
chooses a new node to become the configuration node. The configuration information is
stored in the quorum disks to be safeguarded and accessible by all nodes in the system.
Because the nodes are installed in pairs, each node provides a failover function to its partner
node in the event of a node failure. This failover also enables SAN Volume Controller for
nondisruptive node replacement in case of severe hardware failure, or even for hardware
upgrade to new node models replacing the old ones.
SAN Volume Controller allows temporary coexistence of different node models in the same
pair in those situations, but the normal production environment must have two nodes of the
same model within an I/O group. Different I/O groups in the same system can have distinct
models of nodes.
I/O groups
Each pair of SAN Volume Controller nodes is also referred to as an I/O group. A SAN Volume
Controller clustered system can have from one to four node pairs. A specific volume is always
presented to a host server by a single I/O group of the system for caching purposes with all
the required redundancy and protection. Other I/O groups of the system can be assigned to
the same volume for access purposes to support nondisruptive volume migration between I/O
groups.
When a host server performs I/O to one of its volumes, all the I/Os for a specific volume are
directed to one specific I/O group in the system, or the caching I/O group. Also, under normal
conditions, the I/Os for that specific volume are always processed by the same node within
that I/O group. This node is referred to as the preferred node for this specific volume.
Both nodes of an I/O group act as the preferred node for their own specific subset of the total
number of volumes that the I/O group presents to the host servers. A maximum of 2,048
volumes per caching I/O group is allowed. However, both nodes also act as failover nodes for
their respective partner node within the I/O group. Therefore, a node takes over the I/O
workload from its partner node, if required.
Thus, in an environment based on SAN Volume Controller environment, the I/O handling for a
volume can switch between the two nodes of the I/O group. For this reason, servers that are
connected through Fibre Channel must have paths through the SAN fabrics to both nodes,
and use multipath drivers to be able to handle these failover situations.
If required, host servers can access volumes mapped from more than one I/O group within
the SAN Volume Controller system; therefore, they can access volumes from separate I/O
groups simultaneously. You can move volumes between I/O groups to redistribute the load
between them; however, only selected Operating Systems support this nondisruptive move of
volumes between I/O groups.
Chapter 4. Storage virtualization
95
More information: For more information about the maximum supported objects by I/O
group or system, and also about what operating systems support nondisruptive volume
moves between I/O groups, see the SAN Volume Controller Support Portal web page:
http://www.ibm.com/support/docview.wss?uid=ssg1S1003658
There is a list of supported environments and also a list of the configuration limits for each
version of the SAN Volume Controller software. Use the most appropriate version for the
environment being considered.
System
The system or clustered system consists of one through four I/O groups. Certain
configuration limitations are then set for the entire system or for individual I/O groups. For
example, the maximum number of volumes supported for each system is 8,192 but there is a
maximum of 2,048 volumes for each I/O group. For whichever maximum limitation is reached
first for a specific system configuration, the lowest limitation must be followed.
All configuration, monitoring, and service tasks are performed at the system level, and
configuration settings are replicated to all nodes in the system and saved to the quorum disks
too. To facilitate these tasks, a unique management IP address is set for the system.
A process is provided to back up the system configuration data from the cluster so that it can
be restored in the event of a disaster. This method does not back up application data; only
SAN Volume Controller system configuration information is backed up. For the purposes of
remote data mirroring, two or more systems must form a partnership prior to creating
relationships between mirrored volumes.
MDisks
The IBM SAN Volume Controller system and its I/O groups view the storage that is presented
to the SAN by the back-end controllers as several disks or LUNs, known as managed disks
(MDisks). Because the SAN Volume Controller does not attempt to provide recovery from
physical disk failures within the back-end controllers, an MDisk is usually provisioned from a
RAID array. The only extra protection SAN Volume Controller can have against physical disk
subsystem failures is using Volume Mirroring (for more information about the Volumes
component, see “Volumes” on page 97).
The application servers, however, do not see the MDisks. Instead, they see several logical
disks, known as virtual disks or volumes, which are presented by the SAN Volume Controller
I/O groups through the SAN (Fibre Channel or Fibre Channel over Ethernet) or LAN (iSCSI)
to the servers. The MDisks of same characteristics are then grouped together to form storage
pools where they are divided into several extents, which can be in the range of 16 - 8182 MB
in size, as defined by the SAN Volume Controller administrator.
A volume is a host-accessible unit from storage that has been provisioned out of one storage
pool, or if it is a mirrored volume, out of two different storage pools. A volume is composed by
the extents it takes from the MDisks in the assigned storage pool or pools. The maximum size
of an MDisk is 1 PB and an SAN Volume Controller system supports up to 4096 MDisks
(including those from internal RAID arrays).
Quorum disk
A quorum disk is a managed disk (MDisk) that contains a reserved area of just above 256 MB
for use exclusively by the system to store its configuration and cluster-related data, and also
volume mirroring status information. The remaining area of the MDisks that are selected to
96
IBM SmartCloud Virtual Storage Center
act as quorum disk candidates are still available to provide extents to accommodate volumes
with user data.
The system uses quorum disks for tie-breaking when exactly half the nodes in the system
remain active after a failure, or when a SAN failure occurs, preventing one node from
communicating with the other in the same I/O group. This situation is referred to as “split
brain.”
There are always three candidate quorum disks in a system. However, only one quorum disk
is active at any time. To avoid the possibility of losing all the quorum disk candidates with a
single failure, quorum disk candidates must be assigned on multiple storage systems, if
possible.
For more information about the quorum disk configuration with SAN Volume Controller, see
the following web page in the IBM Knowledge Center:
http://www.ibm.com/support/knowledgecenter/STPVGU_7.3.0/com.ibm.storage.svc.consol
e.730.doc/svc_scsiquorumdiskovr_1bchni.html
Storage pool
A storage pool is a collection of up to 128 MDisks that provides the pool of storage from
which volumes are provisioned. A single system can manage up to 128 storage pools. The
size of these pools can be changed (expanded or shrunk) nondisruptively by adding or
removing MDisks, without taking the storage pool or the volumes offline. At any time, an
MDisk can be a member in only one storage pool.
Each MDisk in the storage pool is divided into several extents. The size of the extent is
selected by the administrator at the creation time of the storage pool and cannot be changed
later. The size of the extent is in the range of 16 - 8192 MB.
A preferred practice is to use the same extent size for all storage pools in a system. This
approach is a prerequisite for supporting volume migration between two storage pools. If the
source and target storage pool extent sizes are not the same, volume mirroring can be used
to migrate volumes between them.
Volumes
Volumes are logical disks that are presented to the host or application servers by the SAN
Volume Controller. The hosts cannot see the MDisks; they can see only the logical volumes
that are created from combining extents from a storage pool.
The three types of volumes are striped, sequential, and image. These types are determined
by the way in which the extents are allocated from the storage pool, as explained here:
򐂰 A volume created in striped mode has extents allocated from each MDisk in the storage
pool in a round-robin fashion.
򐂰 With a sequential mode volume, extents are allocated sequentially from an MDisk.
򐂰 Image mode is a one-to-one mapped extent mode volume.
Using striped mode is the best method to use for most cases because they use extents
coming from all the MDisks presented by a system to the pool, achieving best overall
performance for random workloads to various volumes residing in the same pool. However,
sequential extent allocation mode can slightly increase the sequential performance for certain
workloads with sequential access, by nature. Image mode volumes are normally used to
import preexisting volumes into the virtualized environment, or to export back to native access
from the storage.
Chapter 4. Storage virtualization
97
4.5 Storwize family component overview
The IBM Storwize family of products, which has entry, mid-range, and enterprise members, is
shown in Figure 4-2. Of those members, other than SAN Volume Controller, which is
described in the previous sections, only V5000 and V7000 can participate in the IBM
SmartCloud Virtual Storage Center (VSC) solution. A specific VSC offering is available to be
used with Storwize V5000 and V7000; it is the IBM SmartCloud Virtual Storage Center for
Storwize Family, which aims to have all the VSC functions, but at a lower cost for smaller
environments.
Figure 4-2 IBM Storwize products family and how they relate to VSC offerings
Both Storwize V5000 and V7000 also provide block-level aggregation and volume
management for external attached disk storage systems additionally to their internal capacity.
In this section we briefly explain basic architecture components of Storwize V5000 and V7000
and what is different from SAN Volume Controller. For the characteristics and functions not
mentioned here, they are the same as with SAN Volume Controller.
Note: IBM Storwize V7000 Unified and Storwize V7000 for Flex System are out of the
scope of this Redbooks publication.
Nodes
Each Storwize control enclosure canister is called a node as it pertains to a cluster or system.
A control enclosure has two canisters that communicate with each other through the back
plane. Nodes in one control enclosure communicate to nodes in another enclosure in the
same system through the SAN. This way, the capacity in the expansion enclosures that are
physically connected to one control enclosure through SAS cables becomes available to all
nodes in that system.
External capacity virtualized by a Storwize V5000 or V7000 must be zoned and mapped to all
the canisters in the system. This provides the necessary resources to the virtualization of a
98
IBM SmartCloud Virtual Storage Center
set of volumes, such as cache, processors, and I/O ports. Storwize V7000 Gen2 canisters
include one installed compression accelerator card to help Real-time Compression
processing. A second compression accelerator card is optional.
The same way as SAN Volume Controller nodes are deployed in pairs, or I/O groups, with
Storwize V5000 and V7000, a control enclosure also makes up an I/O group. Multiple control
enclosures may form a Storwize clustered system, or just a system.
I/O groups
Each pair of Storwize V5000 or V7000 canisters within a control enclosure is also referred to
as an I/O group. An I/O group cannot be composed of one node in a Storwize control
enclosure and one node in another control enclosure. Storwize V7000 clusters also allow the
coexistence of different control enclosure generations in the same system.
System
The system or clustered system consists of up to two I/O groups if the system is a
Storwize V5000 or up to four I/O groups if the system is a Storwize V7000.
MDisks
A Storwize V5000 or V7000 system and their I/O groups view both the internal storage
capacity and the capacity that is provided to the SAN by the back-end controllers that they
virtualize. Independent of the source, the capacity is presented to the Storwize system as
several disks or LUNs, known as managed disks (MDisks).
Internal hard disk drives are combined into RAID arrays; each RAID array forms one MDisk.
MDisks of the same characteristics, internal or external, are then grouped together to form
storage pools where they are divided into several extents.
Quorum disk
A quorum disk in a Storwize V5000 or V7000 system is a managed disk (MDisk) with the
same function as in SAN Volume Controller. The difference is that, with Storwize, there is a
choice to use an internal or external MDisk as a quorum disk.
When using internal quorum disks, the availability of a cluster composed by more than one
I/O group can be affected by a remote circumstance of a failure in the control enclosure that
“owns” that MDisk. This failure will make the active quorum disk inaccessible by the other I/O
groups to act as the tie-break. However, all the three quorum disks will still safely hold the
cluster information to resume activity when the system comes up again.
A way to avoid this situation, although unlikely to occur as a result of the lack of a single point
of failure in the Storwize family, is to select one MDisk that is provided by an external
virtualized system as the active quorum disk to support tie-break situations.
For more information about the quorum disk configuration with Storwize V7000 V7.3, see the
following web page in the IBM Knowledge Center:
http://www.ibm.com/support/knowledgecenter/ST3FR7_7.3.0/com.ibm.storwize.v7000.730
.doc/svc_scsiquorumdiskovr_1bchni.html
Chapter 4. Storage virtualization
99
Storage pool
A storage pool in Storwize V5000 and V7000 is also a collection of MDisks. Because internal
and external MDisks might be available, be careful when combining both in the same storage
pool, for example, when the external MDisks are provided from flash technology to compose
an Easy Tier pool together with internal MDisks. In this case, the external system must be
always available, not to affect the entire pool availability. This suggested practice is also valid
for SAN Volume Controller.
Volumes
Volumes are logical disks that are presented to the host or application servers by the Storwize
V5000 or V7000. They are created from combining extents from a storage pool. All the
characteristics of the volumes in SAN Volume Controller also apply to Storwize, including
their use with advanced copy services, mirroring, thin provisioning, Real-time Compression in
the case of V7000, and so on.
4.6 When to choose SAN Volume Controller or Storwize family
When developing an IBM SmartCloud Virtual Storage Center project, there are three
alternatives for the virtualization component of the solution as seen in the previous sections.
Here, we describe the criteria to consider for selecting the appropriate scenario. This is not a
100% comprehensive list, but is a starting point of discussion between the client and the
technical people responsible for the design and sizing of the solution.
Table 4-1 on page 101 compares Storwize V5000, Storwize V7000, and SAN Volume
Controller features when they are part of the VSC offering.
Note: The features and characteristics in Table 4-1are available when each product is part
of the respective VSC offering. This might differ from the standard licensing of the
individual products. For more information about the most current technical specifications of
each product when acquired alone, see the following web pages:
򐂰 For IBM SAN Volume Controller:
http://www.ibm.com/systems/storage/software/virtualization/svc
򐂰 For IBM Storwize V7000:
http://www.ibm.com/storage/storwizev7000
򐂰 For IBM Storwize V5000:
http://www.ibm.com/storage/storwizev5000
100
IBM SmartCloud Virtual Storage Center
Table 4-1 Comparison of Storwize V5000, Storwize V7000, and SAN Volume Controller with VSC
Feature
Storwize V5000
Storwize V7000
SAN Volume Controller
Management
software
IBM Storwize Family Software
for Storwize V5000
IBM Storwize Family Software
for Storwize V7000
IBM Storwize Family Software
for SAN Volume Controller
Internal capacity
Up to 480 drives per control
enclosure (up to 19
expansions); 960 per clustered
system (two-way clustering)
Up to 504 drives per control
enclosure (up to 20
expansions); 1,056 per
clustered system (four-way
clustering)
Up to 48 flash drives per dual
data engines (up to 2
expansions); 192 per clustered
system (four-way clustering)
Port types
1 Gbps iSCSI, 6 Gbps SAS,
8 Gbps FC, 10 Gbps
iSCSI/FCoE
1 Gbps iSCSI, 8 Gbps FC,
16 Gbps FC, 10 Gbps
iSCSI/FCoE
1 Gbps iSCSI, 8 Gbps FC,
16 Gbps FC, 10 Gbps
iSCSI/FCoE
Cache
16 GB per dual controller; up
to 32 GB per clustered system
(two-way clustered)
64 GB or 128 GB per dual
controller; up to 512GB per
clustered system (four-way)
64 GB or 128 GB per dual data
engines; up to 512GB per
clustered system (four-way)
SMP processors
One Intel Xeon E3-1265Lv2
2.5 GHz 4-core processor per
control canister
One Intel Xeon E5-2628Lv2
1.9 GHz 8-core processor per
control canister
One or two Intel Xeon
E5-2650v2 2.6 GHz 8-core
processors per data engine
Clustering
Yes
(up to 2 control enclosures:
4 nodes)
Yes
(up to 4 control enclosures:
8 nodes)
Yes
(up to 4 data engine pairs:
8 nodes)
Compression
No
Yes, Real-time Compression
optional
Yes, Real-time Compression
optional
Unified support
No
NAS connectivity with
Storwize V7000 Unified;
IBM Active Cloud Engine®
integrated
No
Stretched cluster
support
No
No
Yes
Note: For more information about the software functions included in each IBM SmartCloud
Virtual Storage Center offering, see 2.6, “IBM Virtual Storage Center offerings” on
page 51.
When you chose a storage virtualization option, consider these aspects:
򐂰 Performance
Depending on the I/O workload to be processed, consider the equipment that has the
necessary resources, such as cache memory, processors, ports, and the appropriate
number of controllers or nodes to accommodate that workload. The use of a modeling tool
based on the current or expected workload is highly suggested. This will indicate the
minimum system configuration necessary to provide the desired average I/O response
time, and with the hardware resources at acceptable levels of usage for both current and
future needs.
򐂰 Scalability
The initial capacity is an important aspect in any storage project, but also important is to
look at the growth needs for the next years. All three options have their own characteristics
for internal scalability, but worth mentioning is that all of them have external virtualization
capabilities, which expands the possibilities of capacity growth. The use of externally
Chapter 4. Storage virtualization
101
virtualized controllers might affect the overall performance of the system, so they should
be part of the entire sizing study.
򐂰 Particular capabilities
If a particular capability is needed in the project, Table 4-1 on page 101 can help you find
the correct option. For example, if you are considering Real-time Compression, only
Storwize V7000 and the SAN Volume Controller support it. If you require NAS connectivity,
only Storwize V7000 supports Unified File Modules today. And if you are considering a
high available architecture through a stretched cluster, you can have this only with SAN
Volume Controller.
Other aspects can be considered also, such as licensing costs, space and power
consumption, host platform and operating system compatibility, and so on. All of these
aspects must also be considered for storage environments.
102
IBM SmartCloud Virtual Storage Center
5
Chapter 5.
Application-aware data
protection
In this chapter, we introduce the IBM Tivoli Storage FlashCopy Manager product. FlashCopy
Manager is included as part of the license for Virtual Storage Center (VSC) and provides
application-aware data protection through advance snapshot capabilities of storage
subsystems. We provide a high-level overview of the supported FlashCopy Manager
applications, including what application data they can protect.
We also provide a practical use case, which shows how the three products that make up
Virtual Storage Center (SAN Volume Controller, Tivoli Storage Productivity Center, and
FlashCopy Manager for VMware) work together to virtualize, control, and protect your storage
environment.
Notes:
򐂰 This chapter assumes that you have FlashCopy Manager for VMware installed and
configured, which is required for the use case described in 5.3, “Use case: FlashCopy
Manager for VMware custom application support” on page 108. The other assumption
is that you have experience with SAN Volume Controller and Tivoli Storage Productivity
Center.
򐂰 FlashCopy Manager also integrates with Tivoli Storage Manager but requires additional
licenses. Contact your IBM sales representative or IBM Business Partner for assistance
or questions about Tivoli Storage Manager licensing.
© Copyright IBM Corp. 2015. All rights reserved.
103
5.1 FlashCopy Manager overview
IBM Tivoli Storage FlashCopy Manager uses advanced snapshot capabilities of disk
subsystems to create application, database, and VM-aware point-in-time snapshots. These
snapshots can be retained on disk only, or stored by using IBM Tivoli Storage Manager to
take advantage of the full range of advanced data protection and data reduction capabilities.
As the explosion of data continues to increase, and the backup windows decrease, an
application-aware snapshot operation consumes much less time than conventional tape
backup and can be offloaded to a backup server, as shown in Figure 5-1.
IBM Tivoli Storage FlashCopy Manager Overview
Oracle
SAP
SQL Sever
Exchange Sever
Custom Apps
File Systems
VMware
Optional
TSM Backup
Integration
Applications, databases,
virtual systems
Online, near instant snapshot backups
with minimal performance impact
High performance, near instant
restore capability
Integrated with snapshots on a variety
of storage hardware
Database Cloning
Supported Storage Devices
9XIV
9N-series
9DS8000
9NetApp
9FlashSystem V840 9EMC*
9SVC
9HDS*
9Storwize V7000
9Others via VSS
9Storwize V5000
9Storwize V3700
* Via Rocket device adapters
Figure 5-1 Tivoli FlashCopy Manager overview
5.2 Data protection and recovery capabilities
FlashCopy Manager can perform and manage frequent, near-instant, nondisruptive,
application-aware snapshot backups and restores by using the underlying snapshot
capabilities of disk subsystems from IBM and other vendor. Several capabilities are as
follows:
򐂰 Generates application-aware online snapshots in seconds with virtually no backup related
impact on the application server.
򐂰 Can quickly restore from snapshots.
򐂰 Manages multiple snapshots on disk.
򐂰 Can run stand-alone, or optionally off loaded to IBM Tivoli Storage Manager.
򐂰 Does not require a proxy server, with the exception of VMware.
104
IBM SmartCloud Virtual Storage Center
򐂰 Offers Volume Shadow Copy Service (VSS) Instance restore for Microsoft Exchange and
SQL Server.
򐂰 Customer application and fIle system support through use of pre- and post-snapshot
scripts.
򐂰 Offers individual mailbox restore for Microsoft Exchange.
򐂰 Database cloning support.
򐂰 Hardware snapshot backups of VMware VMFS data stores.
򐂰 Supports block-level incremental forever backups with VMware Changed Block Tracking
(CBT).
Note: This applies only when integrating with Tivoli Storage Manager. Exploitation of
CBT is a function of Data Protection for VMware, not of FlashCopy Manager.
򐂰 Restores individual virtual machines from a hardware snapshot of a VMFS data store; no
OS-specific agents are needed in guest.
򐂰 Mount of a backup to a guest for individual drive or files access.
򐂰 Restore of individual virtual disk.
򐂰 Snapshot with remote mirror support.
For a full list of FlashCopy Manager capabilities see the following web page:
http://www.ibm.com/support/docview.wss?uid=swg21427692
5.2.1 FlashCopy Manager: Supported platforms
Some of the supported platforms include Microsoft Exchange, Microsoft SQL Server, and
Windows File Systems using the Microsoft VSS and hardware snapshot providers, DB2, SAP,
Oracle, VMware, and UNIX File Systems.
Depending on the platform, FlashCopy Manager protects various applications, components,
and constructs. FlashCopy Manager for Windows protects Microsoft SQL Server and
Microsoft Exchange. FlashCopy Manager for UNIX protects DB2, Oracle, and SAP.
FlashCopy Manager for VMware protects data stores and the virtual machines inside. Other
applications and components are supported by using custom scripting.
For a complete and current list of supported platforms and versions see this web page:
http://www.ibm.com/support/docview.wss?uid=swg21427692
5.2.2 Components
FlashCopy Manager components can be divided into three distinct categories:
򐂰 FlashCopy Manager for Windows
򐂰 FlashCopy Manager for VMware
򐂰 FlashCopy Manager for UNIX and Linux
An overview of the FlashCopy Manager agents in each of these categories is discussed in the
following sections.
Chapter 5. Application-aware data protection
105
5.2.3 FlashCopy Manager Windows overview
FlashCopy Manager for Windows provides the tools and information needed to create and
manage volume-level snapshots of Microsoft SQL server data, Microsoft Exchange data, file
systems and customer applications. These snapshots are created while the application
remains online. FlashCopy Manager uses the Microsoft Volume Shadow Copy Services,
VSS, to create online point-in-time copies of these applications and file systems and is able to
restore those snapshots to a specific destination. FlashCopy Manager integrates with IBM
hardware and any other hardware that provides a Microsoft VSS provider. Optionally
FlashCopy Manager can be configured for Tivoli Storage Manager support to take advantage
of the full range of advanced data protection and data reduction capabilities.
Protection for Microsoft Exchange
The FlashCopy Manager provides the following protection for Microsoft Exchange:
򐂰 Individual mailbox and mail item restore for Microsoft Exchange servers from a snapshot
backup, including messages, calendars, contacts and other mail objects
򐂰 Support for Microsoft Exchange Database Availability Group
򐂰 Mailbox level restore on a remote system
򐂰 Restore mailboxes directly from the Exchange database files
򐂰 Restoring mailbox messages using the Mailbox Restore Browser
򐂰 Integration with Tivoli Storage Manager (requires additional licensing for IBM Tivoli
Storage Manager for Mail)
Protection for Microsoft SQL Server
The FlashCopy Manager provides the following protection for Microsoft SQL Server:
򐂰 Full and application-aware VSS Microsoft SQL databases backups.
򐂰 Support for Microsoft SQL server cluster environments.
򐂰 Locally stored snapshots for instance restore.
򐂰 Integration with Tivoli Storage Manager (requires additional licensing for IBM Tivoli
Storage Manager for Databases).
Protection for Windows file systems
The FlashCopy Manager provides the following protection for Windows file systems:
򐂰 Microsoft VSS snapshots for Windows NTFS and ReFS.
򐂰 Locally stored snapshots for instance restore.
򐂰 Integration with Tivoli Storage Manager (requires additional licensing).
򐂰 Pre-snapshot and post-snapshot scripts can be configured to quiesce any application to
enable application aware backups.
5.2.4 FlashCopy Manager for VMware
FlashCopy Manager for VMware provides nondisruptive off-host backup, restore, and disaster
recovery solutions for VMware environments. This snapshot approach facilitates faster
backup operations at the virtual machine and data store level. You can restore from
FlashCopy snapshot at the data store, virtual machine, volume, file system and file level.
106
IBM SmartCloud Virtual Storage Center
Protection for data stores and virtual machines
FlashCopy Manager provides the following protection for VMware:
򐂰 Snapshot coverage for VMFS and NFS data stores (NFS data stores must be either IBM
N-Series or NetApp).
򐂰 File level restore granularity from a data-store level snapshot.
򐂰 Ability to exclude virtual machines from data-store level backups.
򐂰 VMware Storage vMotion aware.
򐂰 Support for Virtual machine with virtual disks on multiple data stores
򐂰 VMware template support.
򐂰 Virtual machine backup mode includes VMware snapshot (includes memory), snapshot
exclude memory, suspend virtual machine, and ASIS.
򐂰 Instant restore of multiple data stores.
򐂰 Coexistence with VMware vCenter Site Recovery Manager.
򐂰 Hardware replication between sites through Metro and Global Mirror and remote mirror
snapshot.
򐂰 Integration with Tivoli Storage Manager (requires additional licensing IBM Tivoli Storage
Manager for Virtual Environments).
5.2.5 FlashCopy Manager for UNIX and Linux
FlashCopy Manager for UNIX and Linux provides online backup and restore of data that is
stored in SAP on IBM DB2 and Oracle databases by using the advanced snapshot
technologies of storage subsystems. Also, in combination with IBM DB2 pureScale®,
FlashCopy Manager can back up and restore data on IBM General Parallel File System
(GPFS, also referred to as elastic storage) by using file system snapshots.
Protection for UNIX and Linux
FlashCopy Manager provides the following protection for UNIX and Linux platforms:
򐂰
򐂰
򐂰
򐂰
򐂰
򐂰
DB2 and Oracle with or without SAP
GPFS in combination with DB2 pureScale
Database cloning for DB2 and Oracle
File system and custom application support
Remote mirror snapshot support.
Integration with Tivoli Storage Manager (Requires additional licensing for IBM Tivoli
Storage Manager for Databases)
5.2.6 IBM Tivoli Storage Manager integration
FlashCopy Manager can be integrated with Tivoli Storage Manager by using other Tivoli
Storage Manager products. Depending on the application, Tivoli Storage FlashCopy Manager
can transfer snapshots by using Tivoli Storage Manager for Mail, Tivoli Storage Manager for
Databases, Tivoli Storage Manager Backup-Archive client, or Tivoli Storage Manager for
Virtual Environments; additional licensing is required. These snapshots can be sent to Tivoli
Storage Manager so you can take advantage of the full range of advanced data protection
and data reduction capabilities.
Chapter 5. Application-aware data protection
107
5.3 Use case: FlashCopy Manager for VMware custom
application support
This use case demonstrates how you can use FlashCopy Manager to provide
application-aware data protection for a DB2 database that is running on a VMware virtual
machine.
This use case shows you the process of creating a point-in-time or crash-consistent copy
of a virtual machine that is running a DB2 database. For this case, we use the Tivoli Storage
Productivity Center’s database. To accomplish this, we use FlashCopy Manager for VMware
integration with VMware by using pre-freeze and post-thaw scripts that we created.
Note: The Tivoli FlashCopy Manager DB2 scripts pre-freeze and post-thaw can be
referenced in Appendix A, “Tivoli Storage FlashCopy Manager and DB2 scripts” on
page 227.
Prerequisites
The following prerequisites are for this use case:
򐂰 FlashCopy Manager for VMware must be installed and configured for the storage system
and the vCenter server where the virtual machine is running.
򐂰 VMware tools must be installed in the virtual machine that is running the application you
want to protect.
򐂰 You must write custom pre-freeze and post-thaw scripts for the specific application you are
backing up.
Scripts: An example of pre-freeze and post-thaw scripts is in Appendix A, “Tivoli
Storage FlashCopy Manager and DB2 scripts” on page 227.
򐂰 For Windows 2008 virtual machines that were created using vCenter version before 4.1,
you must enable the disk UUID attribute. More information is at the following web page:
http://pubs.vmware.com/vsphere-50/index.jsp?topic=%2Fcom.vmware.datarecovery.ad
min.doc_20%2FGUID-25A3EBF2-3158-4CB8-A088-133B8EB68120.html
Environment overview
For this example, we configured a small test environment of these components (Figure 5-2 on
page 109):
򐂰 A Windows Server 2008 R2 with VMware vCenter 5.5
򐂰 Two virtual machines running ESXi 5.5
򐂰 A Red Hat Enterprise Linux 6.5 virtual machine with FlashCopy Manager for VMware
installed and configured
򐂰 A Tivoli Storage Productivity Center server, version 5.2.3 virtual machine running on
Windows 2008 R2
򐂰 A IBM SAN Volume Controller version 7.1.0.3 for the back-end volumes for the VMware
data store.These volumes are connected to the ESX servers using ISCSI protocol.
108
IBM SmartCloud Virtual Storage Center
Figure 5-2 Small test environment for FlashCopy Manager for VMware with VSC
Steps for configuring custom application support
Complete the following steps for this use case:
1. Identify which data store the virtual machine that contains the application you want to
backup is running on.
2. Identify which volume that data store represents on your storage subsystem or storage
virtualizer.
3. Create the set of target volumes you plan to use as copies
4. Edit the profile file on the FlashCopy Manager for VMware server, set up or edit the
Device_Class, and set up the Target_Set and either Volumes_File definition or
Target_Name convention.
5. Create the application pre-freeze and post-thaw scripts and place them in the appropriate
directory according to the Table 5-1 on page 110. An example of the scripts is in
Appendix A, “Tivoli Storage FlashCopy Manager and DB2 scripts” on page 227.
Chapter 5. Application-aware data protection
109
Table 5-1 Directory location for freeze and thaw scripts.
Version of ESX
Custom script directory
ESX/ESXi 3.5 Update 1 or earlier
C:\Windows\pre-freeze-script.bat
C:\Windows\post-thaw-script.bat
ESX/ESXi 3.5 Update 2 or later
C:\Program Files\VMware\VMware Tools\backupScripts.d\
ESX/ESXi 4.x or later
C:\Windows\backupScripts.d\
ESXi 5.0
C:\Windows\
C:\Program Files\VMware\VMware Tools\backupScripts.d\
ESXi 5.1 and ESXi 5.5
C:\Windows\pre-freeze-script.bat
C:\Windows\post-thaw-script.bat
Table 5-1 is from the following VMware Knowledge Base web page, which has more
information about pre-freeze and post-thaw script location:
http://kb.vmware.com/selfservice/microsites/search.do?language=en_US&cmd=displayKC
&externalId=1006671
Example steps for this use case
First, we identify the data store on which our target virtual machine is running. To do this, we
log in to the vCenter web, click the virtual machine, click the Summary tab (Figure 5-3), and
look at the Storage field in the Related Objects pane.
Figure 5-3 Identifying the data store the virtual machine is running on.
110
IBM SmartCloud Virtual Storage Center
The next task is to find the mapping between the data store and the volume on the storage
virtualizer. In this example, TPC (Figure 5-5 on page 112) is the name of the data store, and
tpc (Figure 5-4) is the name of the volume.
Figure 5-4 Tivoli Storage Productivity Center shows “TPC” data stores and the VMware esxi server mapping.
Chapter 5. Application-aware data protection
111
Figure 5-5 Using Tivoli Storage Productivity Center, you see the “tpc” volume and the VMware esxi server mapping.
Now that we understand the volume-to-data-store mapping, we can create the target volume
set on the SAN Volume Controller. The target volumes must be at least the same size as the
source volume, in this example the size of the tpc volume is 120 GB. Using this naming
convention helps to more easily identify source-to-target-volumes mapping. The target
volume naming convention used here is tpc_t0, tpc_t1, and tpc_t2 (Figure 5-6).
Figure 5-6 Target volume naming
After the volumes are defined, we use them as the target volumes on the SAN Volume
Controller. We must edit the configuration file on the FlashCopy Manager server, validate the
VMWARE section, and define the DEVICE_CLASS section to be used for this FlashCopy set.
In the VMWARE section, the following parameters must be validated:
򐂰
򐂰
򐂰
򐂰
112
VCENTER_SERVER
AUXILARY_ESX_HOST
VCENTER_SERVER_USER
HOST_NAME_MAPPING.
IBM SmartCloud Virtual Storage Center
In the DEVICE_CLASS section, the following items must be defined:
򐂰 A DEVICE_CLASS for the storage subsystem you use. In this example, we are using a
SAN Volume Controller, we need to define the following parameters, DEVICE_CLASS.
򐂰 COPYSERVICES_HARDWARE_TYPE
򐂰 COPYSERVICES_PRIMARY_SERVERNAME
򐂰 COPYSERVICES_USERNAME
򐂰 COPYSERVICES_REMOTE
򐂰 TARGET_SETS
򐂰 TARGET_NAMING, or VOLUME_FILE, in our example we are using TARGET_NAMING.
The default location for the profile file is /home/tdpvmware/tdpvmware/config.
Several ways exist to set these parameters. You can use a file edit tool like vi and edit the file
directly. If you choose this method you must restart the acs servers manually. You can also
use the setup.sh file, which is in the /opt/tivoli/tsm/tdpvmware/fcm directory. This method
restarts the acs servers for you. The easiest method, new in version Tivoli FlashCopy
Manager version 4.1.1, is to use the configuration wizard in the web interface. Choose the
method that you feel most comfortable using. The text in the boxes must be verified or, for the
DEVICE_CLASS, be created. See Figure 5-7.
Figure 5-7 Example of the profile configuration file on the FlashCopy Manager server
Chapter 5. Application-aware data protection
113
For more information about the configuration file, see the following web page:
http://www.ibm.com/support/knowledgecenter/SS36V9_4.1.1/com.ibm.itsm.fcm.vm.doc/r_
fvm_profileparameters_ovr.html?lang=en
The last step to do before we can run a backup is to create our customer application scripts.
These scripts are in the virtual machine where our customer application is running. In our
case, we create a point-in-time or crash-consistent copy of the DB2 database that is used for
Tivoli Storage Productivity Center. The name of the virtual machine is tpc523.
In our example, we use the process described in the VMware Knowledge Base:
http://kb.vmware.com/selfservice/microsites/search.do?language=en_US&cmd=displayKC
&externalId=1006671
For a detailed review of the scripts in this use case, see Appendix A, “Tivoli Storage
FlashCopy Manager and DB2 scripts” on page 227.
After our scripts are defined and placed in the correct directories, as shown in Table 5-1 on
page 110, we can create a copy of the data store where the virtual machine is located on the
virtual machine running DB2. To accomplish this, we log in to the FlashCopy Manager for
VMware web UI. We define a backup task for the specific virtual machine that we want to
back up, then review the logs to verify that our pre-freeze and post-thaw scripts run.
First, log in to IBM Tivoli Storage FlashCopy Manager for VMware web interface, as shown in
Figure 5-8.
Figure 5-8 FlashCopy Manager for Vmware login page
114
IBM SmartCloud Virtual Storage Center
Next, define a backup task. Do this from the Getting Started window (Figure 5-9).
Figure 5-9 FlashCopy Manager for VMware Getting Started
Click the Define a backup task.
Chapter 5. Application-aware data protection
115
The Create Schedule wizard starts (Figure 5-10) Click Next.
Figure 5-10 The Welcome page in the Create Schedule wizard
116
IBM SmartCloud Virtual Storage Center
The General page opens (Figure 5-11). Provide a name and optional description to the
backup schedule. In this case, the name is DB2. Click Next.
Figure 5-11 The General page in the Create Schedule wizard
Chapter 5. Application-aware data protection
117
The Source panel opens (Figure 5-12). Select the data store and the virtual machines to back
up. In this case, we select the TPC data store and the tpc523 virtual machine.
In most cases, many virtual machines will be on a data store. Although you can choose which
ones you want to back up, they are all available on the target volume. The virtual machines
that are not selected for backup are not available in the Restore panel but the files that make
up those virtual machine are on the target volume. The reason for this is that FlashCopy
Manager provides protection at the volume level.
Figure 5-12 The Source panel is where you select what is going to be backed up
After selecting a source data store and virtual machine, determine the method that you want
VMware to use to create the virtual machine snapshot. Click Next.
118
IBM SmartCloud Virtual Storage Center
The backup settings are listed (Figure 5-13). This is a VMware snapshot and is created
before the hardware snapshot. In this example, we choose to create the VMware snapshot
without memory.
Tip: For more information about these settings, click the Learn more about these backup
settings link in the Backup setting panel.
Click Next.
Figure 5-13 VMware snapshot options.
Chapter 5. Application-aware data protection
119
The Destination settings panel opens (Figure 5-14). In our example, we select SVC (to use
SAN Volume Controller) for our destination device class.
Figure 5-14 Selecting the device class you setup for your target volumes
The backup settings are now defined. Click Next.
120
IBM SmartCloud Virtual Storage Center
The Schedule settings panel opens (Figure 5-15). Here you have the option to schedule this
task for a specific time, or you can execute the task immediately. Click Next.
Figure 5-15 Define a schedule or execute the task immediately
Chapter 5. Application-aware data protection
121
A summary opens (Figure 5-16). This page lists the selections we made in the previous
panels so we can review or choices. You are now ready to execute the task, click Finish.
Figure 5-16 Summary page
After the task is executed, you can view the progress in the FlashCopy Manager UI. Click the
Reports tab under Recent Tasks (Figure 5-17).
Figure 5-17 Recent Tasks report for FlashCopy Manager
122
IBM SmartCloud Virtual Storage Center
We check the log status of our freeze and thaw scripts; we see that both scripts executed with
a return code of 0, as shown in Figure 5-18 and Figure 5-19.
Scripts: Remember, the scripts are in Appendix A, “Tivoli Storage FlashCopy Manager
and DB2 scripts” on page 227.
Figure 5-18 Log for the freeze script
Figure 5-19 Log for thaw script
Chapter 5. Application-aware data protection
123
Look at the vSphere Web Client (Figure 5-20) to see the task that VMware was asked to
execute by FlashCopy Manager for VMware. Here you see that VMware creates a snapshot
of the virtual machine prior to the hardware snapshot.
VMware tools executes our freeze script just before this happens. After the snapshot is
created, the hardware snapshot is created and our thaw script is executed. The VMware
snapshot is then removed.
Figure 5-20 FlashCopy Manager tasks as seen in the vSphere web client.
By returning to the Recent Tasks page (from the FlashCopy Manager UI, click the Reports
tab and then click Recent Tasks), we see that the job completed successfully (Figure 5-21).
Figure 5-21 Successful backup.
To summarize, we accomplished the following tasks:
򐂰 We identified the data store where the virtual machine containing the application that we
wanted to back up was located.
򐂰 We identified the data store to SAN Volume Controller volume mapping.
򐂰 We created a set of target volumes that we used as our targets.
򐂰 We configured the FlashCopy Manager for VMware server, validated the VMWARE
section, set up our Device_Class and Target_Set in the configuration file.
򐂰 We created the application pre-freeze and post-thaw scripts and placed them in the
appropriate directories.
򐂰 We defined a backup schedule, executed it successfully, and validated our freeze and
thaw script logs.
124
IBM SmartCloud Virtual Storage Center
Next, we show the source-volume-to-target relationship in Tivoli Storage Productivity Center.
To do that, we log into our Tivoli Storage Productivity Center web GUI, select Storage
Resources → Storage Systems and then select the specific subsystem we are looking for.
In this example, we specify SVC-2145-JohnWayne-IBM (Figure 5-22). From here we select
Volumes under the Internal Resources list. To simplify our search, we apply a filter to display
only the volumes we are looking for. In this example, we filter on the name tpc.
Figure 5-22 Report shows which volumes are source and which are target
We can dig deeper into this scenario to see the entire FlashCopy relationship. To do this we
look at the properties of the source volume; the source volume is named tpc. Right-click the
tpc volume and select Properties.
The Properties panel opens (Figure 5-23). Click the Relationships tab. Here we can see that
the source volume, tpc has three target relationships: tpc_t0, tpc_t2, and tpc_t2. These are
the target sets we defined in the configuration file for FlashCopy Manager in previous steps.
Figure 5-23 Full source to target volume mapping in Tivoli Storage Productivity Center
Restore options
Next are the restore options with FlashCopy Manager for VMware. Now that we have a
restore-point for our data store and our virtual machine, there are a several methods we can
use for restoring. We can restore a single virtual machine, the entire data store, or attach any
of the virtual machine’s disks from the backup to the running virtual machine.
Chapter 5. Application-aware data protection
125
With the attach feature, we can do a single file-level restore from a full virtual machine-level
backup. To accomplish this, we use the FlashCopy Manager for VMware web GUI. As shown
in Figure 5-24, we first select the Restore and navigate to the virtual machine that we just
backed up (tpc523). Next, we see that we have a restore point and the date and time of that
restore point. From here, we can choose to restore the entire virtual machine or we can attach
one or all of disks to a running virtual machine.
1
2
Figure 5-24 FlashCopy Manager for VMware attachable restore
You can choose your restore point and either restore or attach one or all of the virtual disks
from the backup to a running virtual machine. We choose to attach to a virtual machine, as
shown in Figure 5-25.
Figure 5-25 Attaching virtual disks to a running virtual machine
126
IBM SmartCloud Virtual Storage Center
The attach operation was successful, as shown in Figure 5-26.
Figure 5-26 Recent Tasks shows the attach operation was successful
We can also view the attach operation from the vSphere Web Client. This example shows that
FlashCopy Manager attaches the volume in vSphere, using the original data store name, and
appends the backup ID to it, TPC_A0HSIETQ0M, as shown in Figure 5-27.
Figure 5-27 The entire data store is attached and the backup ID is appended to the data store original
name
Chapter 5. Application-aware data protection
127
We can also view this from a vSphere virtual machine perspective (Figure 5-28).
Figure 5-28 The virtual machine now has the additional disk attached
128
IBM SmartCloud Virtual Storage Center
When we log in to the virtual machine, those disks are listed as additional drives
(Figure 5-29). With this functionality, we can do a single file restore by browsing to any of the
attached volumes and copying any files to be restored.
Figure 5-29 The view from inside the virtual machine
When we no longer need those volumes to be attached, we detach them from the virtual
machine by using the FlashCopy Manager UI. To accomplish this, click the Restore tab
(Figure 5-30), click the virtual machine’s restore point that shows Attached, and then click
Detach. The Detach panel opens. Click Detach.
Figure 5-30 Detach backup volumes from the virtual machine
Chapter 5. Application-aware data protection
129
The volume is removed from vSphere Web Client. See Figure 5-31.
Figure 5-31 vSphere web client view
Notice that the TPC_A0HSIETQ0M volume is no longer available.
Also notice that those volumes were removed from the virtual machine (Figure 5-32).
Figure 5-32 All attached backup volumes are removed without the need to reboot the virtual machine
Instant restore
As demonstrated, we can perform a file system or file level restore based of a virtual machine
or data store level backup. We now do a full virtual machine restore by using the instant
restore function in FlashCopy Manager. In the following steps, we do a data store instant
restore. We delete the virtual machine from the data store, use FlashCopy Manager for
VMware to restore the entire data store, and then log in to the virtual machine and verify that
Tivoli Storage Productivity Center is working and ready for use.
130
IBM SmartCloud Virtual Storage Center
First we delete the virtual machine located on that data store (Figure 5-33).
Figure 5-33 Deleting virtual machine from the data store
Chapter 5. Application-aware data protection
131
Notice that the virtual machine named tpc523 is no longer listed (Figure 5-34).
Figure 5-34 The virtual machine named tpc523 has been deleted.
From the FlashCopy Manager for VMware GUI (Figure 5-35), we click the Restore tab (1),
select Datastores as our restore level (2), and select the most recent restore point (3).
1
2
3
Figure 5-35 Latest data store restore point
132
IBM SmartCloud Virtual Storage Center
The Restore wizard Welcome page opens (Figure 5-36). Click Next.
Figure 5-36 Restore wizard welcome page
Chapter 5. Application-aware data protection
133
We select the virtual machine that we want to register in VMware vSphere. In our example,
we select virtual machine tpc523 (Figure 5-37). Click Next.
Figure 5-37 Virtual machine registration
134
IBM SmartCloud Virtual Storage Center
The Registration Exceptions page opens (Figure 5-38). This page is displayed for several
reasons:
򐂰 If any virtual machines are on the data store that you are about to restore and that are not
in your restore point.
򐂰 It lists any virtual machines that have disk files on other data stores that have not been
selected.
򐂰 It lists any virtual machines that the user did not select for restore.
Click Next.
Figure 5-38 Registration exceptions for VMware vSphere
Chapter 5. Application-aware data protection
135
The Summary page (Figure 5-39) lists several acknowledgement check boxes. These are
warning you that this is a full data store restore, and that virtual machines that are not in the
restore point will be deleted. Click Finish.
Figure 5-39 Summary page
The data store restore is now started. You can view the status by selecting Summary on the
Recent Tasks page. Our instance restore was successful, as shown in Figure 5-40.
Figure 5-40 As you can see from the start and end time, this process took about 2 minutes
136
IBM SmartCloud Virtual Storage Center
The VMware vSphere Web Client shows the virtual machine (tpc523) was restored and is
starting (Figure 5-41).
Figure 5-41 vSphere view of the virtual machine starting up
Chapter 5. Application-aware data protection
137
While the virtual machine is starting, we can look in the SAN Volume Controller GUI at the
running task and see that the volume is still copying the data back to the source volume, or
data store, although the virtual machine is running (Figure 5-42).
Figure 5-42 Flash back operation in progress
138
IBM SmartCloud Virtual Storage Center
The last task is to log in to the virtual machine and issue the restart database command with
the write resume option. To accomplish this, click Start and select IBM DB2 → Command
Line Processor (Figure 5-43).
Figure 5-43 Launch the DB2 command line
From command prompt, issue the restart DB tpcdb write resume command (Figure 5-44).
Figure 5-44 Restart the database to allow writes.
Chapter 5. Application-aware data protection
139
After the command completes successfully, again log in to the Tivoli Storage Productivity
Center server (Figure 5-45).
Figure 5-45 Command completed successfully
You are returned to Tivoli Storage Productivity Center viewing the Copy Relationship reports
(Figure 5-46).
Figure 5-46 Source and target volume report in Tivoli Storage Productivity Center
140
IBM SmartCloud Virtual Storage Center
6
Chapter 6.
Provisioning
Storage provisioning is a known activity for storage administrators. It includes frequent and
repetitive tasks, which must be run in a predefined order to be successful:
򐂰
򐂰
򐂰
򐂰
򐂰
򐂰
Create storage pools
Allocate logical volumes
Create Fabric zones
Define hosts and ports to storage
Assign logical volumes to hosts
Install multipath device driver on host
Although storage provisioning seems a simple process of assigning storage, usually in the
form of server disk drive space, and these processes can be complex. The administrator must
account for both the performance of the newly assigned capacity and the overall storage area
network (SAN) performance. Capacity and performance planning are required to avoid
bottlenecks and lack of resources; a deep knowledge of the technology is needed to be able
explore all its benefits. Doing all of this without Storage Resource Management (SRM)
software can be difficult in medium and large storage environments.
This chapter describes how to use the Provision Storage wizard in IBM SmartCloud Virtual
Storage Center to simplify the way storage is provisioned in an IT environment. Many tasks
are involved when a new request for storage is received by the administrator, and completing
them takes time. With the Provision Storage wizard, the request involves less information and
a shorter amount of time to accomplish and close.
The following topics can help you better understand and use the Provision Storage wizard in
the IBM SmartCloud Virtual Storage Center:
򐂰
򐂰
򐂰
򐂰
򐂰
򐂰
Automated and cloud provisioning
Planning and organizing
Service classes
Capacity pools
Provisioning and sharing volumes
Provisioning storage to cloud and to VMware
© Copyright IBM Corp. 2015. All rights reserved.
141
6.1 Provisioning overview
Several activities must be accomplished when provisioning storage to application servers.
These activities can take from hours to days or even weeks to complete, delaying the
deployment of new applications and overloading the storage administration staff. The
assignment of capacity to servers or virtual machines is a highly manual process, which
requires careful planning for obtaining the best results and not adversely affecting the existing
workloads.
The steps to provision storage must be done in a specific order to be successful. Redundant
data paths must be provided to prevent outages and disaster recovery situations,
performance objectives must be met, and appropriate physical data protection must be
considered. As a result, several areas are susceptible to human errors.
This is where automated storage provisioning, provided by Storage Resource Management
(SRM) software, can help: automating repetitive activities in order to lower the risk of errors,
adding analytics so the characteristics of the environment are used for better decisions, and
reducing the time to process the requests. The storage administration staff then becomes
free for other important activities such as working more proactively with capacity planning, or
evaluating new storage technologies that can benefit their businesses.
6.1.1 Automated provisioning
The best way to avoid errors in repetitive actions is by automating them. This way, a standard
sequence of activities have more predictive results with a higher level of success. The same
concept also applies to the storage provisioning.
Automation can significantly simplify the task of provisioning storage capacity. Each step is
automated and recommended practice rules regarding zoning, device configuration, and path
selection can be applied automatically. The benefits are increased responsiveness to
business requirements, lower administration costs and higher application availability.
With the introduction of a storage services catalog, automation becomes even more
consumable. Now users can request volumes or file shares and receive adequate service
levels based on their profiles. The storage services catalog is created in the IBM SmartCloud
Virtual Storage Center through templates called service classes. For more information see
6.1.6, “Service classes” on page 149.
With the use of the service classes and optionally capacity pools, storage provisioning
requests are run rapidly, assigning capacity to servers with the required quality of service.
The same definitions in the storage environment through IBM SmartCloud Virtual Storage
Center also facilitate the implementation of cloud provisioning, as the next section describes.
6.1.2 Cloud provisioning
Tivoli Storage Productivity Center provides cloud provisioning capabilities through the Cinder
driver for OpenStack environments and the vSphere Web Client extension for VMWare
vSphere. Both methods require the use of service classes.
By defining service classes, the Tivoli Storage Productivity Center administrator can provide
the OpenStack or vSphere administrator with various provisioning options, based on capacity
and also performance and reliability.
142
IBM SmartCloud Virtual Storage Center
Tivoli Storage Productivity Center Cinder driver
With the Tivoli Storage Productivity Center Cinder driver, you can provision block storage by
using Tivoli Storage Productivity Center service classes in an OpenStack environment.
Service classes that are defined in Tivoli Storage Productivity Center are available to
OpenStack as volume types of block storage. For more information about how OpenStack
and Tivoli Storage Productivity Center interact, see 2.5.1, “OpenStack software” on page 41.
When requesting a new volume in OpenStack, and a volume type that corresponds to a Tivoli
Storage Productivity Center service class is selected, the OpenStack Cinder node will send a
bundle of information to the Tivoli Storage Productivity Center server through the Storage
Management API for Clouds (SMAC) API. Information in this bundle includes either or both
worldwide port name (WWPN) and iSCSI initiator information, if available, and also includes
host name, requested storage capacity, and service class information. If only one connection
protocol, Fibre Channel or iSCSI, is available, Tivoli Storage Productivity Center will use only
that connection. If both protocols are available, Tivoli Storage Productivity Center will first try
the protocol specified as preferred_protocol in the cinder.conf file. The default is to use
Fibre Channel first.
Note: The cinder.conf configuration file is placed in the /etc/cinder folder in the server
where the Cinder driver was installed.
Tivoli Storage Productivity Center then uses this information to create a new volume on an
appropriate subsystem, based on the service class definition. After the volume is created
successfully, Tivoli Storage Productivity Center then returns a bundle of information back to
the Cinder node through the SMAC API. The information included in this bundle is volume
identifier information and connection information.
At this point, the OpenStack Cinder node passes this information to the OpenStack Nova
node, and the Nova node will finish attaching the volume, provisioned by Tivoli Storage
Productivity Center, to the OpenStack instance.
Tivoli Storage Productivity Center vSphere Web Client extension
With the Tivoli Storage Productivity Center vSphere Web Client extension, you can provision
block storage in a vSphere environment.
The VMware vSphere Web Client extension communicates with the Tivoli Storage
Productivity Center server through the SMAC API, similar to the Cinder driver. Information
passed to the Tivoli Storage Productivity Center server includes block storage size, service
class name, and capacity pool name.
If capacity is available, the Tivoli Storage Productivity Center server will then create a new
volume on an appropriate subsystem, based on the information provided. After the volume is
created, Tivoli Storage Productivity Center will also assign the volume to the hypervisor or, if a
cluster, all the hypervisors in the cluster through Fibre Channel.
For an example of provisioning with the vSphere Web Client extension, see 6.2, “Use case:
Provisioning by using Tivoli Storage Productivity Center vSphere plug-in” on page 163.
Chapter 6. Provisioning
143
6.1.3 Tivoli Storage Productivity Center: Benefits of using agentless server
The benefits of using the agentless server feature of Tivoli Storage Productivity Center include
provisioning and volume mapping reporting, and volume or application monitoring.
Provisioning and volume mapping reporting
With the agentless server feature in Tivoli Storage Productivity Center, you can provision
volumes and see a view of mapped storage on a remote server without requiring deployment
of a Storage Resource Agent (SRA).
This way is useful in situations where you cannot or do not want to deploy an SRA either for
security restrictions in providing administration credentials, or to avoid loading a production
server with agent code.
Figure 6-1 shows the server named vsc-db02. It was defined to Tivoli Storage Productivity
Center as an agentless server.
Figure 6-1 Properties panel of the agentless server named vsc-db02
144
IBM SmartCloud Virtual Storage Center
Figure 6-2 shows that volumes that are mapped to this server and from which storage
systems, the capacity of those volumes, and the number of paths to those volumes. This
information is useful when troubleshooting performance issues on a specific volume or for a
specific application. By using the agentless feature, you can narrow the search to which
server the volume is attached, and to which application is using that volume.
Figure 6-2 Volumes mapped to a server
The agentless server feature also allows you to provision storage directly to the server and
choose what service class and capacity pool the volume is provisioned from, as shown in
Figure 6-3 and in Figure 6-4 on page 146.
Figure 6-3 Provision storage
Chapter 6. Provisioning
145
Figure 6-4 Volume definition
Volume or application monitoring
The agentless server can be added to your list of monitored servers without deploying an
SRA. This will allow you to view usage and performance of storage that is mapped to that
server as shown in Figure 6-5.
Figure 6-5 Agentless server overview
The agentless server feature can be beneficial for problem determination when an application
experiences performance issues, or to help determine which server or application is driving
high I/O or latency for a specific volume. By defining a server with the agentless feature, you
can quickly determine which volumes are mapped to that server and which volumes are
experiencing performance issues, as shown in Figure 6-6 on page 147.
146
IBM SmartCloud Virtual Storage Center
Figure 6-6 Agentless server volume performance
6.1.4 Planning for provisioning
When planning for storage provisioning, administrators must identify various attributes of their
storage environments, such as these attributes:
򐂰 Existing pools and their unique characteristics (disk technology, RAID type, storage
vendor, and subsystem type)
򐂰 Relative performance between the existing pools in order to assign them to a tier
򐂰 Logical groups of pools to optionally limit the provisioning of capacity from these groups
(capacity pool)
򐂰 User profiles allowed to create provisioning requests, schedule, and run them
򐂰 Desired characteristics of each service class to be created (for example, requires thin
provisioning, uses virtualization, has specific multipath policy for servers, is to perform
automatic zoning, and so on)
The following sections explore these attributes, so administrators can plan accordingly before
implementing automated storage provisioning with IBM SmartCloud Virtual Storage Center.
6.1.5 Organizing your storage environment into tiers
As a prerequisite step for cloud configuration and provisioning, you must assign storage pools
to tiers. You can access a set of panels in Tivoli Storage Productivity Center web GUI, which
explains details of cloud concepts. To access these panels from the navigation pane, select
Advanced Analytics → Provisioning → Learn the concepts.
Storage tiering is a step toward defining service classes and provisioning volumes that
require a certain tier level. It allows you to optimize the placement of volumes. It is also used
in the Tivoli Storage Productivity Center optimization tasks (Analyze Tiering and Balance
Pools).
Chapter 6. Provisioning
147
Considerations for tiering
Carefully consider the organization of your storage environment and allowing for easily
assigning separate categories of data to separate types of storage. One method to facilitate
this process is to initially assign a tier level to all storage pools in your environment.
Although the tiers are used for only SAN Volume Controller and the Storwize family of
products, for the optimization function, provisioning is supported even without SAN Volume
Controller or Storwize products. Therefore, setting up the tiers for all storage pools in your
environment makes sense. Even if you only perform provisioning at the SAN Volume
Controller level, you can still use the tiers of back-end storage pools for reporting, filtering, or
documentation.
Tip: A good practice is to avoid using Tier 10 if possible; most likely, you will not need all
10 tiers. When you start to use reporting, and sort a report by the label, the display will be
similar to the following list, which is most likely not the preferred format:
򐂰
򐂰
򐂰
򐂰
Tier 1
Tier 10
Tier 2
Tier 3
A sample storage categorization is as follows:
򐂰
򐂰
򐂰
򐂰
򐂰
򐂰
򐂰
򐂰
Tier 1. Leave this level available for future uses.
Tier 2. SSD disk pools and flash disks pools
Tier 3. Leave this level available for future uses.
Tier 4. Hybrid pool
Tier 5. Leave this level available for future uses.
Tier 6. FC and SAS disk pools
Tier 7. Leave this level available for future uses.
Tier 8. SATA disk pool
Tivoli Storage Productivity Center V5 has 10 tiering levels
With Tivoli Storage Productivity Center V5, the 10 tiering levels of storage pools are available:
tier 1 (best performing tier) through tier 10 (least performing tier). You can use as many levels
of these tiers as you want, based on the various storage subsystems and disk technologies in
your environment.
In the storage pools tiering process, consider those technologies not present in your storage
environment, but that might be added in the near future. In this way, you are leaving space
among tier levels for assignment later so that you do not need to reconfigure tiers and service
classes if new technologies are added to your environment.
148
IBM SmartCloud Virtual Storage Center
Assigning tiers
To assign a tier level to a storage pool, complete the following tasks:
1. Select Storage Resources → Pools from the navigation pane.
2. Right-click a pool (selecting multiple pools is also possible), select Set Tier, and select a
tier level (Figure 6-7).
Figure 6-7 Storage pool tier assignment
Tips:
򐂰 You can assign a tier to a pool on almost every storage pool table that gets displayed
anywhere in the web GUI.
򐂰 You can filter and sort the table by storage system to easily find the pools where you
want to set up tiers.
6.1.6 Service classes
Service classes enable provisioning automation through infrastructure abstraction, delegating
the determination of the best fit for a storage resource to Tivoli Storage Productivity Center.
Tivoli Storage Productivity Center accounts for all attributes of the service class that are
specified in a provisioning request, and also current storage resource utilization (space,
performance, and status) in order to identify the most appropriate resource for the new
volume or share.
Service classes are a key point in mapping business requirements (capacity, accessibility,
performance, and availability) in infrastructure capabilities (media type, disk technologies,
RAID levels, encryption, compression, and thin provisioning).
Chapter 6. Provisioning
149
Two types of service classes are available:
򐂰 Block-storage service classes
A block-storage service class describes attributes and capabilities of block storage
resources. When you provision volumes, you specify the requirements by using a
block-storage service class. Based on the requirements of the service class, Tivoli
Storage Productivity Center identifies a storage pool for the volume.
Tivoli Storage Productivity Center V5.2 provides the following three predefined
block-storage service classes. Although the classes are predefined, you might need to
change them or adjust your tiers, RAID levels, and so on.
– Gold: Defined for mission-critical applications (highest-performing storage)
– Silver: Defined for applications in production (high-performing storage)
– Bronze: Defined for non-mission-critical applications (standard storage)
򐂰 File-storage service classes
– NormalIsolation: (normal isolation file storage) The file system from which the NAS
share is provisioned can contain other NAS shares. Shared storage is allowed.
– EnhancedIsolation. (enhanced isolation file storage) The file system from which the
NAS share is provisioned cannot contain other NAS shares. Dedicated storage
required.
To determine which storage systems (block and file) are supported for provisioning, see the
following web page. At the web page, select the version you want in the Supported
Products → Storage column:
http://www.ibm.com/support/docview.wss?&uid=swg21386446
Tip: Starting with Tivoli Storage Productivity Center V5.2.1, you can use the candidate
storage tab of the service class dialog to see the matching pools.
To access the Service Classes pane (Figure 6-8), use the navigation menu to select
Advanced Analytics → Cloud Configuration, and then click Work with Service Classes.
Figure 6-8 Service Classes pane
150
IBM SmartCloud Virtual Storage Center
Note: If capacity is indicated in the Unavailable Storage column, this indicates storage
where the status of the pool of storage system has issues, so it will not be used for
provisioning.
From the Service Classes pane, you can view, edit, or delete existing classes. To define a
new service class, click Create Service Class at the top of the Service Class list.
Note: To create a new block-storage service class, the IBM SmartCloud Virtual Storage
Center (VSC) license is required.
Block-storage service classes configuration
From the Service Classes pane, double-click a service class or right-click and select
View/Modify. Figure 6-9 shows the General Service Class properties pane for a
block-storage service class.
Figure 6-9 Block-Storage Service Class Properties pane: General
From this pane, you can customize the service class by editing its attributes.
Chapter 6. Provisioning
151
Table 6-1 describes the General Block-Storage Service Class attributes and the related
resources requirements.
Table 6-1 Block-Storage Service Class General attributes
Attribute
Resource requirements
Storage Tier
If enabled, storage pool must already be assigned to a tier value within the
tier range that is specified in the service class.
RAID level
Storage pool’s RAID level must match (select Any for exclude filtering on a
RAID level).
Virtualization
On: Pool must reside on SAN Volume Controller or V7000.
Off: Pool must not reside on SAN Volume Controller or V7000.
VDisk mirroring
On: Provisioned volumes are mirrored onto a second storage pool.
Off: Provisioned volumes are not mirrored.
(This option is enabled when Virtualization is set to On.)
Thin Provisioning
On: Pools need to allow Thin Provisioning.
Off: None (except for XIV, where you must set Soft Space equal to Capacity
to prevent thin provisioning).
Compression
On: Provisioned volumes are compressed.
Off: Provisioned volumes are not compressed.
(This option is enabled when Thin Provisioning is set to On.a)
Overallocation limit
XIV: Soft Space/Capacity (Soft Space divided by Capacity) must be less
than the specified value.
Other: “Virtual Allocation” must be lower than the specified value.
Encryption
On: Pool must be on a DS8000, encrypted, and configured in the same
Encryption Group as is specified in the service class.
Off: Pool must be not encrypted.b
Redundant Fabric
Checkbox selected. Full redundant paths through a minimum of two fabrics
are used.
Checkbox not selected. All available paths will be used (even single paths).
Resource tags
Storage resources or parent storage system needs to have all specified tags
assigned.
a. Tivoli Storage Productivity Center can create only compressed volumes in I/O groups that
have at least one compressed volume already. If there is no I/O group with a compressed
volume, Tivoli Storage Productivity Center considers this as though compression is not
enabled on the SAN Volume Controller, Storwize V7000, or Storwize V7000 Unified system.
This methodology is used to protect from a system that might have different node hardware;
therefore, be sure to run compression only on certain I/O groups, or in similar situations.
b. At the time of the writing of this book, if encryption is turned on, storage can be provisioned
only directly from a DS8000, which has the corresponding encryption group configured.
152
IBM SmartCloud Virtual Storage Center
Click Advanced to get to the pane for setting advanced properties (Figure 6-10 and
Figure 6-11 on page 154). Set the properties on the pages of these tabs:
򐂰 Thin Provisioning tab
This page of this tab differs in its display depending on the virtualization switch selection in
the General Properties pane (whether virtualization is on or off).
– If virtualization is on (see Figure 6-10)
Note: This display is available only if, on the General Properties pane, the Thin
Provisioning option is set to On.
In this case, the tab indicates SAN Volume Controller and Storwize V7000. From the
page you can set the initially allocated space. If the Auto-expand option is selected,
you can define the granularity of the expansion and the warning level that will trigger a
SAN Volume Controller alert.
Figure 6-10 Thin Provisioning tab: Virtualization On
Chapter 6. Provisioning
153
– If virtualization is off (see Figure 6-11)
In this case, the tab indicates XIV System. You can select the locking behavior of the
pool in case it runs out of space.
Figure 6-11 Thin Provisioning tab: Virtualization Off
򐂰 Multipathing tab
On the page of this tab (Figure 6-12 on page 155), you can select a multipathing policy
(Load balancing, Round robin, or Fail-over), VMware vSphere policy (Default, Round
robin, or Most recently used), and the number of paths that will be defined during
provisioning (Auto, 4, 5, 6, 7, or 8).
Note: If Auto is selected, four paths are configured for IBM storage virtualizer, and two
paths are configured for other storage systems.
Figure 6-12 on page 155 shows the Multipathing page, where the requirement conditions
that must be met to use the multipath policy are listed.
154
IBM SmartCloud Virtual Storage Center
Figure 6-12 Block-Storage Service Class Advanced Properties pane
򐂰 Resource tags tab
Resource tags are used to refine the candidates for provisioning. You can define up to
three resource tags from the tags that are available for selection in the drop-down box, or
you can create your own tags by typing them in the input field. If resource tags are
specified for the service class, only pools that have all the same tags are candidates for
provisioning.
Figure 6-13 on page 156 shows the Resource tags page and some sample tags that can
be used with a block service class.
Chapter 6. Provisioning
155
Figure 6-13 Resource tags available for block service class
File-storage service classes configuration
Figure 6-14 shows the General Service Class properties pane for a file-storage service class.
To access this pane, go to the Service Classes pane, double-click a file-storage service class,
or right-click and select View or Modify.
Figure 6-14 File-Storage Service Class Properties pane: General
From this pane, you can customize the service class by editing its attributes.
156
IBM SmartCloud Virtual Storage Center
Table 6-2 list general file-storage service class attributes and related resource requirements.
Table 6-2 File-storage service class general attributes
Attribute
Resource requirements
Shared storage
Any SONAS or Storwize V7000 Unified file system with free space available,
or unused SONAS Network Shared Disk (NSD) of type “Data, Metadata”.
Dedicated storage
Unused SONAS NSD of type “Data, Metadata”.
Resource tags
Storage resources or parent storage system needs to have all specified tags
assigned.
Click Advanced to open the Advanced Properties pane and the two tabs shown in
Figure 6-15.
Figure 6-15 File-storage service class properties pane: Advanced
Set the properties on the pages of these tabs:
򐂰 General tab
On the page of this tab, you can set these properties:
– Specify whether you want to create an independent file set with its own allocated
inodes. Otherwise, the file set is created as a dependent file set, and is allocated on the
file system. Other considerations to think about with this decision about the file set type
include snapshots, quotas, and Tivoli Storage Productivity Center alerts.
– Choose to set the access path host replacement to the custom tag that specifies the
replacement host name. When the share is provisioned, the access path that is
returned by Tivoli Storage Productivity Center includes the replacement host name, if
one is specified on the custom tag for the storage system. If the storage system does
not specify a value for the custom tag, the cluster name is used in the access path.
Chapter 6. Provisioning
157
򐂰 Resource tags tab
This tab has the same functionality as the Resource tags tab for block-storage service
classes (see “Block-storage service classes configuration” on page 151). Figure 6-16
shows the Resource tags page and some sample tags that can be used with a file service
class.
Figure 6-16 Resource tags available for file service class
Service class and users permissions
Tivoli Storage Productivity Center administrator can grant users the permission to provision
by using the service class. The administrator can also specify whether scheduling or running
provisioning tasks that are created by using the service class requires administrator approval.
To access the service class user permission pane, click Users in the navigation pane of the
service class properties window. In our environment, for example, we defined a user with a
Monitor Role, called TPCoperator, and allowed the user to define a provisioning task on the
Bronze service class. See Figure 6-17 on page 159.
158
IBM SmartCloud Virtual Storage Center
Figure 6-17 Block-Storage Service Class Properties pane: Users
Because the Approval required properties is set to Yes, the TPCoperator user can only
define and save the task. An administrator can later schedule or run it.
Note: If Approval required is set to No, the defined user setting for TPCoperator, in
addition to its monitoring role, can define and save the task, and also schedule or run it.
Chapter 6. Provisioning
159
Optional: Associate capacity pools with the service class
A service class can be optionally associated with one or more capacity pools. See 6.1.7,
“Capacity pools” on page 161 for details about capacity pools.
If capacity pools are defined, the Storage Constraints entry will be available in the navigation
pane of the Service Class window. See Figure 6-18.
Figure 6-18 Storage Constraints pane
160
IBM SmartCloud Virtual Storage Center
Candidate storage
Storage pools that satisfy all requirements of a service class are listed in the Candidate
Storage pane (Figure 6-19). All criteria must be satisfied or the storage pool will not be listed.
In the example, test_service_class was created with the requirement of storage tier 10 at
any RAID level. Only the volumes that met all criteria are displayed.
Figure 6-19 Service class candidate storage
6.1.7 Capacity pools
A capacity pool is a logical grouping of the following resources:
򐂰
򐂰
򐂰
򐂰
Storage systems
Storage pools
File systems of file storage systems
IBM Scale Out Network Attached Storage (SONAS) Network Shared Disks (NSDs)
Organizing storage resources in separate capacity pools in any way that serves the required
business needs is possible. For example, it can be used to separate storage resources that
are installed in geographically separate sites, or that must be allocated for separate divisions
(and also for separate customers).
By organizing resources in capacity pools, you can track the storage use for each division
separately, and restrict provisioning requests to the appropriate set of storage resources, by
restricting the service classes to specific capacity pools.
Chapter 6. Provisioning
161
Optional: The use of capacity pools is optional.
To access the Capacity Pools pane, complete the following steps:
1. From the navigation menu, select Advanced Analytics → Cloud Configuration.
2. Click Work with Capacity Pools as shown in Figure 6-20.
From this pane, you can select the available resources in your environment that you can
assign to a capacity pool, by clicking the corresponding link.
Figure 6-20 Capacity Pools pane
To assign a resource to a capacity pool, complete the following steps:
1. Go to the resource pane, right-click the resource, and select Add to Capacity Pool.
2. If your environment does not have any capacity pools, the Create Capacity Pool pane
opens (Figure 6-21).
Figure 6-21 Create Capacity Pool pane
162
IBM SmartCloud Virtual Storage Center
3. In the Create Capacity Pool pane, you define a new capacity pool. enter a name for the
capacity pool and optionally a description and up to three tags.
When you add a storage system to a capacity pool, note that any storage pool, file system,
or IBM SONAS NSD that is an internal resource of that storage system is also indirectly
assigned to the capacity pool.
Restriction: A resource can be assigned to only one capacity pool. When you attempt
to add more resources to a capacity pool, and those resources are already assigned to
a different capacity pool, Tivoli Storage Productivity Center displays a message that
lists the resources and their capacity pool assignments. When a message is displayed,
ensure that you want to change the capacity pool assignments for the listed resources.
4. If capacity pools are already defined, the Adding Resources pane opens (Figure 6-22).
Select a capacity pool and click Save.
Figure 6-22 Adding Resources pane
6.2 Use case: Provisioning by using Tivoli Storage Productivity
Center vSphere plug-in
This use case describes an example of how Virtual Storage Center (VSC) can be used so
that the VMware administrator can provision storage without requiring intervention from the
storage administrator.
This use case demonstrates how a VMware administrator can provision storage by using the
Tivoli Storage Productivity Center vSphere plug-in with minimal interaction of the storage
administrator. This is not intended to be a detailed setup process.
Chapter 6. Provisioning
163
We completed the following steps in preparation for this use case:
򐂰 Configured a data source for the VMware vSphere environment, either a single ESX
server or VMware vSphere server, to the Tivoli Storage Productivity Center server.
򐂰 Configured the data source for the storage subsystem. In this case we are using SAN
Volume Controller, which was defined to the Tivoli Storage Productivity Center Tivoli
Storage Productivity Center server.
򐂰 Created a user ID on the Tivoli Storage Productivity Center server that we used to define
the Tivoli Storage Productivity Center vSphere plug-in to the vSphere web GUI. This user
ID was also authorized to allow provisioning permissions when we defined the service
class on the Tivoli Storage Productivity Center server.
򐂰 Created a service class and assigned capacity pools on the Tivoli Storage Productivity
Center server.
򐂰 Registered the Tivoli Storage Productivity Center vSphere plug-in on the vSphere server.
For details about the steps, see the following web page:
http://www.ibm.com/support/knowledgecenter/SSNE44_5.2.4/com.ibm.tpc_V524.doc/fqz0_
t_vm_deploying_extension_locally.html?lang=en
We used the following steps:
1. Configure the VMware vSphere environment to the Tivoli Storage Productivity Center
server, as shown in Figure 6-23. We use the oden ESX server name throughout this case.
Figure 6-23 VMware ESX servers as shown in Tivoli Storage Productivity Center GUI
2. Define the data source for the storage subsystem that is being used for the back-end
storage for our VMware data stores. For this case, we use a Storwize V7000 named
Storwize V7000-2076-v7000-storea-IBM (Figure 6-24 on page 165).
164
IBM SmartCloud Virtual Storage Center
Figure 6-24 Storage subsystem data source as shown in Tivoli Storage Productivity Center
3. Define a group and user account on the Tivoli Storage Productivity Center server that we
use for our service class, then we add the group in the Tivoli Storage Productivity Center
GUI. For this case, the group name is vmware and the role that is assigned to the group is
External application, as shown in Figure 6-25.
Figure 6-25 User Management in Tivoli Storage Productivity Center
4. Define vmware as the capacity pool name (1 in Figure 6-26) and assign storage (2) to it.
1
2
Figure 6-26 Capacity poll properties page
Chapter 6. Provisioning
165
5. After the capacity pool is defined, create a service class named vmware (Figure 6-27).
Figure 6-27 Service class general properties panel
6. Assign the capacity pool to the service class (Figure 6-28).
Figure 6-28 Capacity pool named vmware assigned to service class named vmware
166
IBM SmartCloud Virtual Storage Center
7. We also define the user with provisioning permission to this service class (Figure 6-29).
Figure 6-29 Restricting service class access to a specific user
8. After we configure all the items on the Tivoli Storage Productivity Center server, we
register the Tivoli Storage Productivity Center vSphere plug-in on the VMware vSphere
server (Figure 6-30).
Figure 6-30 The plug-in registration
Chapter 6. Provisioning
167
9. After the plug-in is registered, we log in to the VMware vSphere Web Client and save the
user ID and password we want the vSphere server to use to connect to the Tivoli Storage
Productivity Center server for the provisioning tasks (Figure 6-31).
Figure 6-31 Tivoli Storage Productivity Center user ID defined in the vSphere web client
10.Now that our Tivoli Storage Productivity Center server and our vSphere Web Client are
configured, we are able to provision storage from the vSphere Web Client, which is based
on the service class, capacity pool, and user permissions we just defined. In this use case,
we provision a 5 GB volume to the ESX server named oden. To accomplish this, we
right-click on oden, and then select All TPC Actions → Provision Block Storage, as
shown in Figure 6-32 on page 169.
Note: In this type of scenario, only the designated storage pool can be provisioned.
Even if more storage enclosures or pools are available in the same V7000, those
cannot be managed by the Tivoli Storage Productivity Center vSphere plug-in.
168
IBM SmartCloud Virtual Storage Center
Figure 6-32 Provision action in the vSphere web client
11.This action opens a provisioning pane, in this example Provision Block Storage. This is
where you can choose the size for the volume, service class, and capacity pool you want
to provision storage from. In this use case, we give vmware user access to only one service
class and one capacity pool (Figure 6-33).
Figure 6-33 Provisioning a volume and creating a data store
Chapter 6. Provisioning
169
12.After executing the task, we look in the Tivoli Storage Productivity Center GUI to see that
the provisioning task is running (Figure 6-34).
Figure 6-34 Running provisioning task as seen in Tivoli Storage Productivity Center
13.In several minutes, the provisioning task is complete. We can now view the newly created
volume and data store in the vSphere web GUI (Figure 6-35).
Figure 6-35 Newly created 5 GiB data store.
In this use case, we demonstrate how customers can provision a volume and create a data
store by using the integration that provided by the Tivoli Storage Productivity Center vSphere
plug-in. We also show how to restrict who has access to provision storage and what class of
storage we can restrict the users to.
Note: In addition to the Tivoli Storage Productivity Center vSphere plug-in, the vStorage
API for Array Integration (VAAI) is also supported for provisioning storage.
170
IBM SmartCloud Virtual Storage Center
7
Chapter 7.
Storage optimization
In this chapter, we describe storage optimization functions that are available with IBM
SmartCloud Virtual Storage Center (VSC). Storage optimization functions help you to
optimize your storage environment by improving performance and better utilizing storage
resources.
Storage optimization functions use real performance data of the managed storage
environment. By using this data, relative activity score of storage resources is calculated.
According to the activity score, Tivoli Storage Productivity Center, the storage management
interface, proposes recommendations made by VSC, which you can than run in your
virtualized environment.
© Copyright IBM Corp. 2015. All rights reserved.
171
7.1 Storage optimization functions and processes
Storage optimization functions in VSC are part of advanced storage analytics, which help you
to improve performance of your storage environment and to better use storage resources. By
using the advanced analytics, storage optimization functions can do these tasks:
򐂰 Storage optimization: Balance the workload of volumes across storage pools
򐂰 Volume optimization: Move volumes from a higher tier to a lower tier and vice versa
򐂰 Volume transformation: Transform volumes, for example, convert volumes from fully
allocated to thin-provisioned or compressed, or vice versa.
Storage optimization functions work on a volume level in a virtualized environment where the
scope of optimization functions are within one storage virtualizer (SAN Volume Controller
cluster or Storwize V7000). If you have one storage virtualizer in your environment that is
running production systems and another storage virtualizer that is running test and
development systems, optimization is done for each storage virtualizer separately.
Several steps are required before you can use storage optimization functions to optimize an
environment:
򐂰 Set the tier level of the storage pools that you want to analyze.
򐂰 Probe the storage virtualizers.
򐂰 Collect performance data to gauge the utilization percentage for pools and the average
workload activity of the volumes.
Note: For more details about advanced analytics and storage optimization, see IBM Tivoli
Storage Productivity Center V5.2 Release Guide, SG24-8204.
Setting tier levels
Tivoli Storage Productivity Center provides 10 tier levels that can be assigned to storage
pools. For storage optimization functions you need to set only a tier for storage pools of
storage virtualizers (SAN Volume Controller or Storwize V7000). Even if you are not using
optimization functions, a good practice is to use tier levels; tiering allows you also to group,
sort, and filter similar storage pools and create reports based on the tier level.
Setting tier levels is done in Tivoli Storage Productivity Center web GUI from navigation pane.
To assign a tier level to a pool, see “Assigning tiers” on page 149.
Probing and collecting data for storage virtualizers
Storage optimization functions require regular data collections for each storage virtualizer. By
default, when you add a device through the Configure Device Wizard, you configure a time for
probe and performance collections. Probe jobs run against each device to update the asset
level details known about each device. Performance jobs run against each device to collect
performance details of the asset within the array. You can also modify the start and stop times
for these jobs on the details page for the array.
To check data collection jobs for storage virtualizer, complete the following tasks:
1. Select Storage Resources → Storage Systems from the navigation pane.
2. Double click the Storage Systems for which you want to check data collection jobs.
3. From Storage System details pane, click Data Collection; the Data Collection pane
shows details of the probe and performance monitor jobs (Figure 7-1 on page 173).
172
IBM SmartCloud Virtual Storage Center
Figure 7-1 Data Collection jobs
If probe jobs are not running for the selected resource, select Actions → Start Probe
(Figure 7-1) to manually start a probe. The Start Probe action is available for the resources
that you select if at least one of the resources meets the following criteria:
򐂰 A probe job is defined for the resource
򐂰 The probe job is enabled
򐂰 The probe job is not currently running
Probes are data collection jobs; they collect status and asset information about the monitored
resources in your environment. When you add a storage resource to Tivoli Storage
Productivity Center, a probe is automatically defined to run to collect status and asset
information about the monitored resource. You can also schedule the probe job to run when
appropriate for your environment.
Note: Probe jobs must run at least every day, but we suggest to start them manually before
optimization tasks to be sure that you collect the most current information about your
storage environment. If some changes occur in your storage resources, they will be
collected with the probe job.
Collecting and analyzing performance data
Performance monitors collect data about storage virtualizer pools to provide a realistic
estimate of the average workload activity of the volumes. When you use the Analyze Tiering
wizard to analyze the tiering of volumes, performance data must be collected about the pools
on the storage virtualizers that you select for analysis, and about the storage virtualizer pools
that you select as target pools for the volumes that are analyzed. The collected performance
data is used to determine the workload activity of the pools. The most active and least active
pools are identified by comparing the activity of each pool to the average activity of the pools
on the same tier and in the same storage system.
Chapter 7. Storage optimization
173
To determine the relative activity of the pools on the same tier and on the same back-end
storage system, the following values are calculated:
򐂰 The current activity level of the pool
򐂰 The activity deviation percentage of the pool
Tip: On the Pools page, the activity level of the pool is shown in the Activity column. The
difference between the activity level of the pool and all of the other pools on the same tier
and storage system is shown in the Activity Deviation (%) column.
The formulas that are listed in Table 7-1 calculate the activity level and the activity deviation
percentage for pools.
Table 7-1 Formulas for calculating the activity level and activity deviation percentage values for pools
Value calculated
Formula that is used in the calculations
Activity level of the pool
[Read I/O Rate × (1- Read I/O Cache Hit%)] ÷ Total Pool
Capacity
Activity level of the pool
on XIV Systems
(Total I/O Rate ÷ Total Capacity)
Activity deviation
percentage of the pool
[(Pool Activity Level - Average Activity Level) ÷ Average
Activity Level] × 100
򐂰
򐂰
A positive value indicates that the relative activity level of the pool is
above average.
A negative value indicates that the relative activity level of the pool is
below average.
If the activity deviation percentage of the pool is in the range of <-10% to >10%, the pool is
considered to be balanced. If the activity deviation percentage of the pool exceeds the
deviation threshold of 10%, the pool is a candidate for balancing.
The minimum period for collecting performance data for storage optimization functions is one
day but the preferred period for collecting performance data is 14 days or more. The period for
collecting the performance data can provide an accurate reflection of the regular workload
activity of the volume.
If, for example, the collected performance data covers a period of extreme workload activity,
the average that is estimated does not reflect the regular workload activity of the volumes.
Insufficient performance data, or performance data that does not reflect the regular workload
activity of a volume, might affect the accuracy of the recommendations that are provided by
the analysis.
To better understand how the storage optimization process is done and how analyzing tiers
and balance pools functions work, see Figure 7-2 on page 175. It shows a sample storage
environment that consists of four tiers and nine storage pools.
Capacity pools: Use of capacity pools is optional and is not shown in Figure 7-2. A
capacity pool is a logical grouping of the following resources:
򐂰
򐂰
򐂰
򐂰
174
Storage systems
Storage pools
File systems of file storage systems
IBM SONAS Network Shared Disks (NSDs)
IBM SmartCloud Virtual Storage Center
Figure 7-2 Storage optimization
The balance pools function is used to balance the workload of volumes across pools on the
same tier. The pools are analyzed and recommendations are generated to move the volumes
from pools with high-activity values to pools with low-activity values. For example in
Figure 7-2 if you are balancing pools in Tier 3, a recommendation could be to move some
volumes from Pool 3-1 and Pool 3-3 if the workload activity on those pool is higher than on
Pool 3-2. In this example workload will be distributed within same tier.
The analyze tiering function provides a simple and unified optimization of volume placement
in storage environment. For example, as shown in Figure 7-2, you might try to first balance
the workload across Tier 2 to distribute the workload evenly across all Tier 2 pools. If however
no balancing is required, you might consider to analyze tiering at the volume level to move
volumes from Tier 2 up or down a tier based on the current activity of the volumes or pools.
The volume transformation function is also an optimization function. It is used to move
volumes between pools and to convert volumes from fully allocated to thin-provisioned or
compressed volumes, or vice versa.
Chapter 7. Storage optimization
175
The storage optimization process is done in the following steps:
1. Select resources for optimization.
The first phase of the storage optimization process is to select the resources. You can
initiate the optimization tasks from many places in the Tivoli Storage Productivity Center
web-based GUI either through the actions menus or when you right click an object.
Depending on the type of object you selected (storage system, pool, volume, or server)
you might not see all functions. For example, if you just recently selected volumes, the
balance pool function is not available.
These are the contexts from which you can start the analysis:
–
–
–
–
–
Server (either with SRA or Agentless)
Hypervisors
Storage systems
Pools
Volumes
With the exception of some panels in the Advanced Analytics menu, you can start the
optimization from any panel that lists volumes or pools.
2. Define policy and specify parameters.
This part is specific to the type of optimization that you are running. We documented this
for the two functions separately here:
– 7.2, “Balance Pools” on page 178
– 7.3, “Analyze Tiering” on page 185
3. Run the analysis.
The analysis first checks some prerequisites and then opens a wizard that displays
informational messages. You can enter information such as thresholds. When you finish,
you click Analyze. An analysis task is then created and runs in the background. If you do
need to do other work, you can close the dialog, and return to that task later. The task is
available in the context of the devices that are related to the task, and also on the Task tab
of the Storage Systems and Tasks panel within the Home menu.
After the analysis completes and the results are displayed, you can change the default
name of the analysis task to a more meaningful name. If you closed the wizard, you can
open the results of the analysis task from one of the task panels. From the results of the
analysis task, you can also take the action to schedule the analysis part to run repeatedly
every scheduled number of days.
4. Review the recommendations.
Review the results of the analysis task. Either the wizard is still open, or you can go to a
task panel, and double-click an analysis task to review the results.
5. Execute recommendation or schedule the execution to run later
If the recommendations are suitable, you can let Tivoli Storage Productivity Center
implement them. Click Execute. Otherwise, you can set up a one-time scheduling task by
selecting Schedule → Execution. Because this is a one-time task, it cannot be repeated.
The execution of the task is based on the VDisk mirror functions. Tivoli Storage
Productivity Center is actually adding a VDisk copy, synchronizing it, and removing the
initial copy (in the same way that Tivoli Storage Productivity Center is doing for volume
transformations). While the execution is running, you can pause, cancel, or resume the
task.
176
IBM SmartCloud Virtual Storage Center
Storage optimization process for all three functions is described in the following sections:
򐂰 7.2, “Balance Pools” on page 178
򐂰 7.3, “Analyze Tiering” on page 185
򐂰 7.4, “Volume transformation” on page 197
Easy Tier considerations with Tivoli Storage Productivity Center
optimization and provisioning
The main difference between storage optimization and Easy Tier is that storage optimization
works on volume level, and Easy Tier works on extent level.
Easy Tier monitors the host I/O activity and latency on the extents of all volumes with the
Easy Tier function turned on in a multitier storage pool over a 24-hour period. It then creates
an extent migration plan that is based on this activity, and it dynamically moves high activity or
hot extents to a higher tier within the storage pool. It also moves extents whose activity
dropped off, or cooled, from the high tier back to a lower tier. Because this migration works at
the extent level and not at the volume level, it is often referred to as sub-LUN migration.
Storage optimization monitors the activity and workload of all volumes and pools. By
analyzing the performance data of storage environment, it moves volumes with low workloads
from higher and more expensive tiers to lower and less expensive tiers. By moving the
volumes it optimizes the storage environment.
The Easy Tier function is suggested if you want to put a volume in an Easy Tier pool, if only a
portion of its extents are hot and the remainder of the volume is fairly cold. If you have
volumes that are typically hot or cold throughout the extents, you can consider using storage
optimization to move them up or down a tier.
Consider the following information regarding Easy Tier usage and Tivoli Storage Productivity
Center optimization and provisioning:
򐂰 Easy Tier does not provide any initial volume placement options; Tivoli Storage
Productivity Center provisioning does.
򐂰 Easy Tier does use an exponential moving average during its calculation, so it does look
at more than a day, but earlier performance monitor statistics get “reduced” in their
importance.
򐂰 Easy Tier does not actually try to improve volume performance, but it tries to avoid volume
hot spots. The result is that some volumes get faster and others not, but there is no real
influence that you can enforce. For example, a test volume might consume much of the
SSD and flash capacity, and leave little to a production volume.
򐂰 Because Easy Tier might optimize a volume based on the heat per extent, a situation can
happen in which the Tivoli Storage Productivity Center assumption about the IOPS that
are distributed across the complete volume might be wrong and might lead to wrong
recommendations.
Chapter 7. Storage optimization
177
Table 7-2 compares SAN Volume Controller Easy Tier and Analyze Tiering function.
Table 7-2 SAN Volume Controller Easy Tier versus Analyze Tiering
Characteristic
SAN Volume Controller Easy Tier
Analyze Tiering
Purpose
Find and optimize hottest extents.
Move workload between pools.
Optimization strategy
Avoids creation of hot spots on
MDisks, based on “knowledge” of
microtiers.
Acts only if a large variation of the
response times (skew) exists.
Multiple Strategies:
򐂰 Default: move volumes out of
tier/pool if activity limit is
exceeded.
򐂰 Advanced: depending on policy
settings (I/O rate or I/O density,
and age).
Analysis
Monitors cumulative response time
per MDisk (in the same pool).
Runs continuously.
Monitors activity score per pool (in
the same tier) over multiple days
On-demand or scheduled.
Granularity
Extent.
Volume. But analysis can be
initiated for storage system, pools,
volumes, servers, or hypervisors.
Boundary/Type of pool
Hybrid pools (tiering within pools
also called microtiering).
One or more tiers (each tier can be
multiple pools).
Analysis time frame
Cannot be adjusted.
Can be adjusted.
Automation/Control
Completely automatic.
Semi-automated tool for lifecycle
management.
Range of tiers is determined by
service class.
Initial tier placement
Depends on available capacity and
cannot be controlled.
Tier determined by service class
settings and current workload.
Move/Swap
One tier at a time.
Volume can “jump” across multiple
tiers.
Life Cycle Management
No
Yes
Also included
Balancing of MDisks within pool.
Balancing workload across multiple
pools of the same tier level.
7.2 Balance Pools
The Balance Pools function (wizard) is used to balance the workload of volumes across
storage pools in the same tier. It analyzes the pools performance, and according to
recommendations done on pool activity, volumes are moved from pools with high-activity
values to pools with low-activity values. With this option, a more balanced workload is
achieved across the pools within same tier.
Balance Pools function is supported with the following storage systems:
򐂰
򐂰
򐂰
򐂰
178
IBM SAN Volume Controller
IBM Storwize V7000
IBM Storwize V7000 Unified
IBM Storwize V5000
IBM SmartCloud Virtual Storage Center
Important:
򐂰 Any storage that is supported by SAN Volume Controller or Storwize, and that is
configured as a storage pool (managed disk group) in the storage system, can be used
with this function. This is because the function works on the SAN Volume Controller or
Storwize level.
򐂰 Collecting performance data for the back-end storage systems (which can be also a
storage system not from IBM) is unnecessary because the performance data is
collected on the storage virtualizers.
Considerations for using balance pools
Consider this information when you use the Balance Pools function:
򐂰 You select pools, not volumes (like in Analyze Tiering), and you must select at least two
pools per storage system, per tier.
򐂰 The analysis is always based on the primary copy of a volume or VDisk, because that is
the volume receiving the read I/Os.
򐂰 If you have a multiple-site configuration (SAN Volume Controller stretched cluster), be
careful when you select the target pools for this optimization so that Tivoli Storage
Productivity Center will not move the volumes of a local server into another location. Doing
this can result in a higher response time between storage and server, although the
back-end is better utilized. Therefore do not mix pools of different locations.
Note: You can always use the pools custom tags to specify a location manually so that
you can easily filter when you select pools from a list of pools.
򐂰 The Balance Pools function differs from the SAN Volume Controller Perl rebalance script
as shown in Table 7-3. There are Scripting Tools for SAN Volume Controller including a
Perl script for rebalancing of the virtualized storage across the underlying storage, which
can be downloaded from this site (requires an IBM ID for downloading):
http://ibm.co/1c9h0Gr
Table 7-3 Comparing Balance Pools and re-balance script
Characteristic
Tivoli Storage Productivity
Balance Pools
SAN Volume Controller
Perl re-balance script
Boundary
Across pools
One pool
Purpose
Analyze and optimize volume
performance
Avoid hotspots simply by re-striping extents
across all MDIsks, without considering actual
performance numbers.
Granularity
Volumes
Extents
Tip: You can use the SAN Volume Controller Perl re-balance script with older SAN Volume
Controller versions. Since SAN Volume Controller version 7.3, the balancing is done
automatically based on the performance rather than only the capacity and there is no need
to use the script.
Chapter 7. Storage optimization
179
Steps for balancing pools
To use the Balance Pools function (wizard) and redistribute volume workload across pools on
the same tier, complete the following steps:
1. Check that the prerequisites are in place:
– Probe has run and data about storage virtualizer is collected.
– Performance monitor is running and data is collected.
– Storage pools are assigned to a tier.
2. Select two or more pools on the same tier.
3. Define policy and specify parameters for the analysis. In the policy, complete these steps:
– Include or exclude volumes in mirrored volume relationships.
– Specify the number of days that are used to estimate the activity levels of the pools.
– Specify whether you want to restrict the placement of volumes in capacity pools to
storage pools in the same capacity pool. This option is available only if one or more of
the pools that are being analyzed belong to capacity pools.
– Specify whether volumes that are assigned to the same hypervisor or server must be
placed in the same destination pool.
– Optional: Click Advanced balancing to set an activity limit for the pools on each tier
4. Run the analysis.
A recommendation is generated according to the defined policy.
5. Review the recommendation.
6. Start or schedule the recommended task.
Details about the steps are in 7.5, “Use case: Balance Pools” on page 200.
When to use Balance Pools: Three scenarios
Here are three situations where you might want to use the Balance Pools function.
Scenario 1
You recently provisioned new volumes to your server from Pool 2 in Tier 2 and performance
data shows that the pools within this tier are not balanced. The performance data is shown as
activity level of the pools, which is calculated with the formula in Table 7-1 on page 174.
Because you are running more of a workload on Pool 2 (activity level is much higher on
Pool 2, than on Pool 3), Tier 2 is likely to be used in an unbalanced way (Figure 7-3 on
page 181).
180
IBM SmartCloud Virtual Storage Center
Figure 7-3 Scenario 1: Balance Pools
The Balance Pools function first analyzes the pools in Tier 2 (Figure 7-4).
Figure 7-4 Scenario 1: Balance Pools analysis
According to the analysis, the Balance Pools function recommends which volumes should be
migrated so that each pool in the tier receives approximately the same workload, and thus
avoids hot spots.
Chapter 7. Storage optimization
181
After you review the recommendation, start or schedule the execution task (Figure 7-5), which
will move volumes from Pool 2 to Pool 3. The virtual disk copy function is used to move the
volumes, the workload is not disrupted.
Figure 7-5 Scenario 1: Balance Pools execution task
After the optimization execution task completes, volumes are moved and the storage pools
will be balanced (Figure 7-6).
Figure 7-6 Scenario 1: Balance Pools completed
182
IBM SmartCloud Virtual Storage Center
Scenario 2
You recently added a new storage pool (Pool 4) to your environment and it was set up and
assigned in the same tier as your existing storage pool (Figure 7-7).
Figure 7-7 Scenario 2: Balance Pools
Because the workload is running only on Pool 2 and Pool 3, Tier 2 is likely to be used in an
unbalanced way. The Balance Pools function first analyzes the pools (Figure 7-8).
Figure 7-8 Scenario 2: Balance Pools analysis
Chapter 7. Storage optimization
183
According to the analysis, the Balance Pools function recommends which volumes should be
migrated so that each pool in the tier receives roughly the same workload, and thus avoids
hot spots. After you review the recommendation, start or schedule the execution task
(Figure 7-9), which will balance pools and move volumes from Pool 2 and Pool 3 to Pool 4.
The virtual disk copy function is used to move the volumes and the workload is not disrupted.
Figure 7-9 Scenario 2: Balance Pools execution task
After the optimization execution task completes, volumes are moved and the storage pools
will be balanced (Figure 7-10 on page 185).
184
IBM SmartCloud Virtual Storage Center
Figure 7-10 Scenario 2: Balance Pools completed
Scenario 3
As workloads change over time, you can let Tivoli Storage Productivity Center run a balance
analysis on a set schedule and see if there is potential to optimize the distribution of the
volumes. According to recommendations, you can start or schedule the execution task.
7.3 Analyze Tiering
The Analyze Tiering function (wizard) is used to optimize the placement of volumes on
storage tiers. Based on the performance data, information about back-end storage
technology and conditions that are set in tiering policies, volumes are analyzed and
recommendations are generated. After you review the recommendations, you can execute
the recommendations to move the volumes on the tiers that best match the workload
requirements.
Introduction
When Analyze Tiering is used, these concerns are considered by functions and by the user:
򐂰 Performance capability constraints can exist. Even if a storage pool has sufficient physical
capacity for many more storage volumes, the additional I/O load of these volumes might
be such that one or more of the storage components (for example, the backing
subsystem) becomes overwhelmed. This results in degraded performance for all storage
volumes in the storage pool. Tivoli Storage Productivity Center allows the specification on
an Activity Limit to help avoid such situations.
򐂰 Although multiple pools in the same tier are likely to be using the same (or similar) disk
technology, the actual storage pool performance scales with the number of disk drives.
Tivoli Storage Productivity Center’s activity score measure normalizes non-cache I/O
activity by storage pool capacity as a basic approximation for this.
Chapter 7. Storage optimization
185
When you use Analyze Tiering functions, Tivoli Storage Productivity Center looks at the
workload of the selected volumes or pools (and the storage pools that the selected volumes
belong to), and also the current workload and the capability of the target pools that were
specified in the analysis task wizard. Tivoli Storage Productivity Center will run an analysis
and provide recommendations for the movement of volumes, if it can find any better
placement.
The default and advanced options
You can use Analyze Tiering function with default or advanced options:
򐂰 Default options
With default options, a balance algorithm is used, which will move volumes to a different
tier (if available) if the current tier cannot be balanced.
򐂰 Advanced options
With advanced options, you can fine-tune your tiering and act, even before limits are
reached, by trying to find a more optimal placement for a volume. You do this by providing
minimum thresholds for I/O rate or I/O density, and file usage that volumes need to meet in
order to qualify to be placed in a certain tier.
Calculations: To better align with the activity score calculation, the I/O rate of a volume is
determined by retrieving, over the specified time range, the volume’s average daily read
I/O, write I/O, and read cache hit percentage and then calculating:
write IO rate + read IO rate * (1 – read cache hit %)
This same I/O rate calculation is used for the numerator when determining a volume’s I/O
density:
IO Rate / capacity in GiB
Use default options if you do not know what the good values are for the advanced options
thresholds. If you have pools, volumes, and applications that require special handling or
should never be included in an analysis, consider using the advanced options.
For both options you specify an activity limit for the destination pools. Volumes are not added
to destination pools if they cause the activity level of the pools to exceed the activity limit that
you want to specify for pools. It is a contingency value that you can specify as an upper
boundary so that you do not overload a pool, by moving volumes into a target pool.
Working with the default and advanced options
If you start the analysis by selecting several volumes, Tivoli Storage Productivity Center will
determine which pools they belong to. If there is more than one pool per tier, the default
options will be used to balance pools before making any up or down tiering suggestions. The
process is as follows:
1. Starting with the lowest tier of storage, the source pools are balanced.
2. If the pools cannot be balanced, the most-active volumes are moved to a higher tier of
storage. This step does not apply to pools on the highest tier of storage.
3. The source pools are balanced again.
4. If the pools cannot be balanced, the least-active volumes are moved to a lower tier.This
step does not apply to pools on the lowest tier of storage.
5. The source pools are balanced again.
6. Steps 1 - 5 are repeated until the highest tier is balanced.
186
IBM SmartCloud Virtual Storage Center
Regardless which option you choose to optimize storage, you must set an activity limit for
each tier of storage. Activity limits are set to maintain the performance capability of the
destination pools. If you are unsure what activity-limit value to use, consider the Tivoli Storage
Productivity Center reported maximum, and average activity values in relation to your
understanding of your storage device’s performance. If no performance problems exist, set
the activity limit somewhat higher than either the maximum or average value (for example,
20). If performance problems do exist, set the activity limit somewhat lower than either the
maximum or average value (for example, -20). If you are unsure, set the activity limit equal to
the current maximum value.
How Analyze Tiering works
Here are the basic functions of Analyze Tiering:
򐂰 You can analyze tiering by these components:
– Servers: Analyze tiering of volumes on storage virtualizers that are connected to
servers
– Hypervisors: Analyze tiering of volumes on storage virtualizers that are connected to
hypervisors
– Storage virtualizers: Analyze tiering of volumes on storage virtualizers
– Storage pools: Analyze tiering of volumes on storage pools
– Volumes: Analyze tiering of volumes
򐂰 Analysis starts at the lowest tier (tier 10) through to the highest tier (tier 1).
򐂰 Analysis always tries to optimize the storage within a tier level first by balancing pools,
before making a recommendation to move a volume up or down a tier. You might find
recommendations to move a volume within a tier.
򐂰 Tivoli Storage Productivity Center looks only at the selected pools or volumes and the
specified targets. It never makes a recommendation outside of the selected source and
target pools.
򐂰 No performance data from the back-end storage systems is required for the optimization.
򐂰 From the specified target pools, Tivoli Storage Productivity Center works only with pools
that meet the following conditions:
– The pool must be online.
– The pool must be in a non-error state.
– The pool must have sufficient available space.
򐂰 If you schedule the optimization run to be executed at a later time, Tivoli Storage
Productivity Center does the following basic checks, but does not run the analysis again:
– Check for sufficient space.
– Check that the volume is still in the same source pool.
– If the volume is using VDisk mirror, the status is checked.
򐂰 During analysis, the activity score of a destination pool must not exceed the activity limit
value.
򐂰 Default and advanced options:
– Default options: During the analysis, Tivoli Storage Productivity Center uses the new
calculated average and maximum activity for all storage pools within the same tier. In
addition, a desired activity limit must be calculated based on the sum of the maximum
read I/O rate and write I/O rate operations that you want to specify as limits, divided by
the average capacity of all the pools on the same tier.
Chapter 7. Storage optimization
187
– Advanced options: In addition to the data from the Automatic mode, I/O density or I/O
rate with minimum threshold level and percentage of file accessed and the period of
accessing the files are used for tiering analysis.
򐂰 Tivoli Storage Productivity Center is using the VDisk mirror function to move a volume
from one pool to another. This is online and transparent to the application, and can even
be used if the pools are using different extent sizes.
򐂰 For volumes that are already using VDisk mirroring, the primary copy gets analyzed.
When you execute a recommendation for a mirrored volume, Tivoli Storage Productivity
Center will delete the primary copy and create a new copy in the recommended location.
You can choose which of the volume copies will be set as the primary copy after
completing the execution. For more information see 4.2.3, “Enhanced data availability” on
page 92.
Volume collocation
The Analyze Tiering function minimizes the exposure of servers or hypervisors to multiple
back-end storage systems by collocating volumes that are assigned to the same hypervisor
or server.
For example when you want to analyze the volumes that are in storage pools on a Storwize
V7000 storage virtualizer with different back-end storage systems, be sure that the volumes
in the same storage pool that are assigned to the same server or hypervisor are kept
together. With the Analyze Tiering option, you enforce the collocation of volumes and you
prevent the placement of related volumes in destination pools that might be on multiple
back-end storage systems.
In the Analyze Tiering wizard and the Balance Pools wizard, you can enforce the collocation
of volumes that are in the same source pool, and that are assigned to the same hypervisor or
server.
Additional suggestions
Before you implement the Analyze Tiering function, also consider this information:
򐂰 Tiering Analysis can be set up in a way that you can use capacity pools as boundaries for
recommendations. You may also choose whether to enforce service class restrictions.
򐂰 Although moving volumes across virtualization clusters is not possible, you can select
multiple virtualization clusters in one analysis task.
򐂰 Be aware that Tivoli Storage Productivity Center Tiering Analysis looks at only
performance and not availability. Therefore, when you select pools as target pools, be sure
sure that they provide the same level of availability as the source pools.
If you consider one pool to be more available than another pool, you can use the capacity
pool concept to group the pools. You can then define to only make recommendations
within the boundaries of capacity pools.
򐂰 If you are using SAN Volume Controller Easy Tier pool, and Tivoli Storage Productivity
Center recommends to move a volume from a SAN Volume Controller Easy Tier pool
down to a pool that is not an Easy Tier pool, check the SAN Volume Controller heat map
by using the Storage Tiering Advisor Tool (STAT) to confirm that this volume can really be
moved down. The I/O density or the file age calculation can sometimes lead to a decision
that is not optimal.
򐂰 Tivoli Storage Productivity Center uses Secure Shell (SSH) sessions to communicate with
the SAN Volume Controller or Storwize storage system. The number of available SSH
sessions are limited to ten with recent versions of the SAN Volume Controller software. As
a result, do not run too many tasks in parallel, or Tivoli Storage Productivity Center might
use all the sessions.
188
IBM SmartCloud Virtual Storage Center
SAN Volume Controller stretched cluster considerations
Consider the following information when you use an SAN Volume Controller stretched cluster
configuration.
򐂰 Volume movement is implemented by using SAN Volume Controller VDisk mirroring. If you
are using SAN Volume Controller VDisk mirroring in stretched cluster environments, Tivoli
Storage Productivity Center must drop one of the VDisk copies first, before it can
implement any recommendation. You can argue that this might compromise the availability
of a volume, but SAN Volume Controller limit is to have a maximum of two VDisk copies
per volume.
򐂰 The optimization wizard asks if you want to include mirrored volumes.
򐂰 Tivoli Storage Productivity Center does not currently analyze and optimize both copies.
Tivoli Storage Productivity Center always deletes the primary copy of a volume, and
creates a new copy in the recommended pool. During the definition of the optimization
task, you can choose which of the volume copies should act as the primary copy after the
task is completed:
– If the new copy will act as the primary copy, your application can benefit from the
optimization, but the other copy (which stays the secondary copy) is left in the original
tier.
– If the secondary becomes the primary, your application does not immediately benefit
from the optimization, but now this copy receives the real workload. It is s likely to be
optimized in the next optimization run.
򐂰 You can use capacity pools to group the storage pool for each site even if you do not plan
to use any cloud configuration. This enables you to limit Tivoli Storage Productivity
Centers analysis to make only up and down tiering suggestions within one site. You can
still use a single job for the analysis.
Important: If you use the capacity pools to group the storage pool for each site, you might
find that during provisioning, Tivoli Storage Productivity Center will not create the volume
copies in two separate locations. That is because the Tivoli Storage Productivity Center is
lacking a site concept and what you define for optimization might not work for provisioning.
Considerations for using Analyze Tiering
Consider the following information for using Analyze Tiering:
򐂰 Tivoli Storage Productivity Center is always moving full volumes from one tier to the other,
so Tivoli Storage Productivity Center tiering is not the same as the SAN Volume Controller
and Storwize Easy Tier function.
Note: All movement of storage will be contained within the same storage system.
򐂰 The analysis is always based on the volume copy with preferred read because that is the
volume receiving the read I/Os.
򐂰 The analysis is a two-phase task:
a. In the first phase, you select the resources and parameters, and Tivoli Storage
Productivity Center will run an analysis. Because analyzing the resources can take
time, Tivoli Storage Productivity Center also provides scheduling capabilities.
b. In the second phase, review and execute the recommendations where Tivoli Storage
Productivity Center can actually implement the recommendations for you; you can do
this immediately or create a schedule to be run at a more convenient time. Always
review the recommendations before you execute them.
Chapter 7. Storage optimization
189
Steps for analyzing tiers
To analyze tiers and move the volume workload across tiers, complete the following steps:
1. Check that the prerequisites are in place:
– Probe has run and data about storage virtualizer is collected.
– Performance monitor is running and data is collected.
– Storage pools are assigned to a tier.
2. Select volumes or pools for tiering analysis and determine the placement of optimized
volumes by selecting one or more target pools.
3. Define a policy where you choose how you want to tier storage.
If you want to tier volumes based on the workload requirements of the volumes, you can
set minimum thresholds for each tier of storage. You can specify minimum thresholds for
I/O density or I/O rate, and file usage. You can also combine minimum thresholds for I/O
density and file usage, and for I/O rate and file usage. The workload of the volumes is
evaluated to determine whether the volumes require re-tiering. If you want to balance
pools and restrict the re-tiering of volumes to pools that cannot be balanced across each
tier, you do not specify tiering criteria. You must also set an activity limit for each tier of
storage.
4. Run the analysis.
A recommendation is generated according to defined policy on selected pools and
volumes.
5. Review the recommendation.
6. Start or schedule the recommended task.
Details about the steps are described in 7.6, “Use case: Analyze Tiering” on page 210.
When to use Analyze Tiering: Three scenarios
Here are three situations when you might want to use the tiering analysis.
Scenario 1
Over time, workloads might change or the server that is requesting volumes might have
overestimated or underestimated requirements, so the volumes can no longer be located in
the optimal storage pool (Figure 7-11 on page 191).
190
IBM SmartCloud Virtual Storage Center
Figure 7-11 Scenario 1: Analyze Tiering
In this scenario, you want to perform Analyze Tiering for the two volumes in Pool 2. The
analysis in this scenario strives to optimize the storage within a Tier 2 level first by balancing
pools, before making a recommendation to move a volume up or down a tier. You start the
Analyze Tiering function for two volumes in Pool 2 (Figure 7-12).
Figure 7-12 Scenario 1: Analyze Tiering analysis
Chapter 7. Storage optimization
191
According to the analysis, the Analyze Tiering function recommends where the volumes
should be migrated to in order to balance the workload. After you review the recommendation,
start or schedule the execution task (Figure 7-13), which in this scenario is to move volumes
from Pool 2 to Pool 3 and balance the pools. The VDisk copy function is used to move the
volumes and the workload is not disrupted.
Figure 7-13 Scenario 1: Analyze Tiering redistributing volumes
After the execution task completes, volumes are moved to Pool 3 and the workload is running
normally and balanced (Figure 7-14).
Figure 7-14 scenario 1: Analyze Tiering completed
192
IBM SmartCloud Virtual Storage Center
Scenario 2
If you have a performance problem for one set of volumes, you can use this function to
determine whether to move those volumes to a higher tier level. In this example, two volumes
have performance problems (Figure 7-15).
Figure 7-15 scenario 2: Analyze Tiering
To determine the placement of these two volumes, select target pools and set thresholds
such as I/O rate or I/O density to Analyze Tiering (Figure 7-16).
Figure 7-16 Scenario 2: Analyze Tiering analysis
Chapter 7. Storage optimization
193
According to the analysis, the Analyze Tiering function recommends moving the volumes to a
higher tier. After you review the recommendation, start or schedule the execution task
(Figure 7-17). The recommendation in this scenario is to move volumes from Tier 2 to Tier 1
and improve performance of selected volumes. The VDisk copy function is used to move the
volumes, and the workload is not disrupted.
Figure 7-17 Scenario 2: Analyze Tiering, moving volumes to higher tier
After the execution task completes, volumes are moved to Tier 1 and the workload runs
normally (Figure 7-18 on page 195).
194
IBM SmartCloud Virtual Storage Center
Figure 7-18 Scenario 2: Analyze Tiering completed
Scenario 3
Perhaps you allocated volumes on a high performance tier because there was free capacity
at that time (Figure 7-19). Over time, the tier can fill up, so you must continuously check to
determine if volumes should be moved to another tier. In this scenario, you can set up Tivoli
Storage Productivity Center so that the analysis phase runs regularly (for example, every one
to 14 days).
Figure 7-19 Scenario 3: Analyze Tiering
Chapter 7. Storage optimization
195
After the analysis runs (Figure 7-20), where you selected Tier 2 pools and Tier 3 pools as a
target pools, a recommendation is provided about moving the volumes to a lower tier.
Figure 7-20 Scenario 3: Analyze Tiering analysis
After you review the recommendation, start or schedule the execution task (Figure 7-21),
which in this scenario moves volumes from Tier 1 to Tier 3 and releases Tier 1 from additional
performance. The VDisk copy function is used to move the volumes and the workload is not
disrupted.
Figure 7-21 Scenario 3: Analyze Tiering, moving volumes to lower tier
196
IBM SmartCloud Virtual Storage Center
After the execution task completes, volumes are moved to Tier 3 and the workload runs
normally (Figure 7-22).
Figure 7-22 Scenario 3: Analyze Tiering done
7.4 Volume transformation
The volume transformation function in Tivoli Storage Productivity Center is used to convert
volumes in storage pools to fully allocated volumes, compressed volumes, or thin-provisioned
volumes. It can be also used to move volumes to other pools on the same storage system or
to pools that are enabled for Easy Tier on the same storage system.
Note: Tivoli Storage Productivity Center uses VDisk copy to implement most of the
transformations. This means that volume transformation is supported with only IBM SAN
Volume Controller, IBM Storwize V7000, and Storwize IBM V7000 Unified.
The following volume transformation options can be used:
򐂰 Migrate pool
Migrate pool moves the selected volumes from a pool to those pools that you selected
from the available pools. This option can be used if you want to move some volumes to a
different pool, which is on another back-end storage system or on a pool that has slower
disk drives. Also you can bring a new storage system and move all the volumes from an
old storage system to the new system. In that way, you can remove the old storage
system.
򐂰 Compression
Compression is used to convert your fully allocated volume to a compressed volume, or to
transform a compressed volume to a fully allocated volume. This option provides
immediate capacity savings in your storage environment.
Chapter 7. Storage optimization
197
IBM storage virtualizers are using the Random Access Compression Engine (RACE)
technology, which compresses data that is dynamically written into the storage system.
This compression occurs transparently, so Fibre Channel and iSCSI connected hosts are
not aware of the compression. RACE is an inline compression technology, meaning that
each host-write is compressed as it passes through the storage virtualizer software to the
disks. This technology provides immediate capacity savings and it can also be used for
primary storage workloads, such as virtual machines, databases, and active data set
applications.
򐂰 Thin provisioning
Thin provisioning enables the storage to present the required capacity to the host while
allocating only the actual used capacity in terms of space on the physical storage media.
By using the thin provisioning function, you can convert fully allocated volumes to
thin-provisioned volumes or thin-provisioned volumes to fully allocated volumes.
By transforming volumes to thin-provisioned volumes, you can better utilize your storage
environment and provide some capacity savings. In some cases, you might want to
transform thin-provisioned volumes to fully allocated volumes. This might be a case where
you used some thin-provisioned volumes for test, and now you want them to use for
production where all the volumes are fully allocated volumes.
򐂰 Easy Tier
By using the Easy Tier feature in the volume transformation function, you are moving
selected volumes from non Easy Tier pool to Easy Tier pool to benefit from Easy Tier. Non
Easy Tier pools are usually pools with high-performing or low-performing storage media;
Easy Tier pools are with mixed storage media (high-performing and low-performing).
Easy Tier is a performance function that automatically migrates or removes extents from a
volume to, or from, one MDisk storage tier to another MDisk storage tier. Easy Tier
monitors the host I/O activity and latency on the extents of all volumes with the Easy Tier
function turned on in a multitier storage pool over a 24-hour period (heatmap creation).
Next, it creates an extent migration plan based on this activity and then dynamically moves
high activity or hot extents to a higher disk tier within the storage pool. It also moves
extents whose activity has dropped off or cooled from the high-tier MDisks back to a
lower-tiered MDisk.
Multiple choices
The volume transformation function allows multiple choices. For example, you might want to
implement compression or thin provisioning on a volume and move the volume in a different
pool at the same time. The following options are possible:
򐂰
򐂰
򐂰
򐂰
򐂰
Migrate pool + thin provisioning
Migrate pool + Easy Tier
Migrate pool + thin provisioning + Easy Tier
Migrate pool + compression + thin provisioning + Easy Tier
Compression + thin provisioning
Multiple choices are also available. For example if you want to implement compression or thin
provisioning on a volume and move the volume in a different pool at the same time, you select
both options and create the task.
How volume transformation works
The volume transformation functions are based on the VDisk copy feature, which is supported
only for volumes that are defined within storage virtualizers: SAN Volume Controller, Storwize
V7000, and the Storwize V7000 Unified storage systems.
198
IBM SmartCloud Virtual Storage Center
When you initiate the volume transformation function in Tivoli Storage Productivity Center,
it will start VDisk copy task in storage virtualizer for a selected volume. The VDisk copy
task will create a copy of the selected volume and start the synchronization process. The
synchronization process updates the copy until it is fully synchronized. When the copy is fully
synchronized, the storage virtualizer will split off a synchronized volume copy to a new
volume and the workload will continue on the new transformed volume. The original primary
volume is deleted after the transformation function is completed.
Transformation functions require temporary extra space to convert volumes to
thin-provisioned or fully allocated volumes. Also when migrating pools, additional
capacity is required in target pools.
Some volume transformation functions depend on other functions. So, for example, when you
select Easy Tier, the Migrate Pool function is automatically selected.
Volume transformation restrictions
Restrictions apply when you implement volume transformation. The restrictions for converting
or moving volumes are as follows:
򐂰 To move volumes, the destination pool must be on the same storage virtualizer. You
cannot move a volume from a pool on one storage virtualizer to a pool on another storage
virtualizer.
򐂰 To convert or move volumes, image mode volumes must be converted to managed mode
volumes.
򐂰 To convert fully allocated volumes to compressed volumes, you must have the IBM
Real-time Compression license.
򐂰 Only Tivoli Storage Productivity Center, administrators can implement the Transform
Storage functions.
When to use volume transformation
You can use the volume transformation functions to change certain characteristics of
selected volumes to improve performance or optimize the use of available storage. Volume
transformation functions can also be used if you want to migrate data from one back-end
storage to another in case you are replacing old back-end storage with a new one.
Volume transformation steps
To run the volume transformation functions and change characteristics of your volumes,
complete the following steps:
1. Select one or more volumes from Tivoli Storage Productivity Center web GUI
2. For the selected volumes, select a Transform Storage option and select volume settings
that you want to change. You may select more than one option.
3. Enter additional information for the selected Transform Storage options.
4. The Transform Storage function will create the task.
5. Start or schedule the transformation.
Details about the steps are described in 7.7, “Use case: Volume transformation” on page 219.
Note: You can find more details about volume transformation in IBM Tivoli Storage
Productivity Center V5.2 Release Guide, SG24-8204.
Chapter 7. Storage optimization
199
7.5 Use case: Balance Pools
You can use the Balance Pools function (wizard) to redistribute workloads across pools in
same tier.
You may start the Balance Pools function from any list of pools. In this use case, we start the
Balance Pools function from the Tivoli Storage Productivity Center navigation pane
(Figure 7-23) by selecting Advanced Analytics → Optimization.
Figure 7-23 Tivoli Storage Productivity Center navigation pane
From the optimization pane, select Optimize Pools to see all your pools. We select only the
pools that belong to one storage virtualizer (svc1prod), and we select the pools from Tier 3.
The activity deviation percentage shows that the pools are not balanced (Figure 7-24).
Figure 7-24 Optimize Pools pane
Either select Actions → Balance Pools or right-click the selected pools and select Balance
Pools (Figure 7-25 on page 201).
200
IBM SmartCloud Virtual Storage Center
Figure 7-25 Optimize Pools pane: Balance Pools function
After Tivoli Storage Productivity Center checks prerequisites, the Balance Pools window
opens, where you define a policy and specify parameters for the balance pools analysis.
(Figure 7-26).
Figure 7-26 Balance Pools policy
In this use case we analyze performance data over the previous 30 days. Because we defined
only one capacity pool for all pools of our storage virtualizatior, we select No for the
placement of optimized volumes in capacity pools. We also select No to collocate volumes
because the volumes in selected pools are used only for test. By selecting No we do not
restrict volumes that are assigned to the same server or hypervisor from being separated
between the source and destination pool.
We also set the pool activity limit for the destination pools, which prevents adding volumes in
the destination pools if their workload will exceed that limit.
Click Analyze to start the Balance Pools analysis for the selected pools and defined policy.
Note: During the analysis phase, Tivoli Storage Productivity Center looks at pools with an
above average utilization and analyzes the volumes. If any pools fit the criteria, they should
be moved to other pools in the same tier. The target pools must be able to accept the
additional workload so that the balance can work; if all pools have a higher utilization than
you specified in the Balance Pools wizard, Tivoli Storage Productivity Center will not be
able to optimize anything.
Chapter 7. Storage optimization
201
Balance analysis successfully completed (Figure 7-27). The activity of the pools can be
balanced.
Figure 7-27 Balance Pools analysis
The Recommendation page (Figure 7-28) shows which volumes will be moved. In this case,
the TPC_SVC1_0001 and TPC_SVC1_0002 volumes will be moved from the XIV_OTHER pool to the
XIV_BERNARD pool.
Click the Open Logs link.
Figure 7-28 Balance Pools recommendations
202
IBM SmartCloud Virtual Storage Center
The Logs pane opens, which shows the log of the Balance Pools analysis (Figure 7-29).
Figure 7-29 Balance Pools analysis log
The volumes that will be moved to the new pool are mapped to the test Windows server
IICTPCRT1-SEC (Figure 7-30). You see that the volumes are in the XIV_OTHER pool.
Figure 7-30 Test server volumes
Chapter 7. Storage optimization
203
Figure 7-31 shows the Disk Management window on the Windows server where you see the
same volumes.
Figure 7-31 Disk Management of the test server
Figure 7-32 shows that we are running a workload on those volumes, by using an I/O meter
tool.
Figure 7-32 Workload on the test server
204
IBM SmartCloud Virtual Storage Center
The same workload is shown in the storage virtualizer performance monitor (Figure 7-33).
Because this workload is the only one running on this storage virtualizer, several IOPS are
almost the same.
Figure 7-33 Workload on the test server shown in SAN Volume Controller GUI
After we review the recommendation (Figure 7-28 on page 202), we schedule the Balance
Analysis task (Figure 7-34).
Figure 7-34 Balance Analysis task schedule
Chapter 7. Storage optimization
205
Figure 7-35 shows all storage systems tasks and the Balance Pools task that we scheduled.
Figure 7-35 Storage Systems tasks status: scheduled
When the task starts, the status changes to Running (Figure 7-36).
Figure 7-36 Storage Systems tasks status: running
Also when the task starts, the storage virtualizer moves the volumes to the new pool. It uses
VDisk copy to move the volumes. The VDisk copy task creates a copy of selected volumes
and starts the synchronization process.
206
IBM SmartCloud Virtual Storage Center
Figure 7-37 shows a VDisk copy task in the SAN Volume Controller GUI; it shows that the
new copies of selected volumes (Copy 1) are configured in a new pool.
Figure 7-37 VDisk copy
The synchronization process updates the copy until it is fully synchronized. Figure 7-38
shows the synchronization progress in the SAN Volume Controller GUI.
Figure 7-38 Synchronization progress
When the copy is fully synchronized, the storage virtualizer will split off a synchronized
volume copy to a new volume and workload will continue on the new volume. The original
primary volume is deleted after the Balance Pools function completes. Figure 7-39 on
page 208 shows the SAN Volume Controller GUI and only the new volumes in the new pool.
Chapter 7. Storage optimization
207
Figure 7-39 Volume status in SAN Volume Controller GUI
The Storage Systems tasks pane (Figure 7-40) shows that the Balance Pools task
successfully completed.
Figure 7-40 Successful completion of Storage Systems tasks
208
IBM SmartCloud Virtual Storage Center
If you open the log, it shows all the steps that are done (Figure 7-41).
Figure 7-41 Balance Pools log
If you check the volumes for the Windows server, you see that they are now in the new pool
XIV_BERNARD (Figure 7-42).
Figure 7-42 Volumes status in Tivoli Storage Productivity Center web GUI
The most important objective of this use case is to show that the workload is now balanced
across pools as shown in Figure 7-43 on page 210. Before we did the balancing, the
XIV_BERNARD pool did not have any workload, and the XIV_OTHER pool was fully utilized.
After we balanced the pools, you see that the workload is now running on both pools
according to the defined policy.
Chapter 7. Storage optimization
209
Figure 7-43 Pools performance
7.6 Use case: Analyze Tiering
This case shows how to use the Analyze Tiering function (wizard) at the volume level by
setting minimum thresholds for I/O rate on each tier. The Analyze Tiering function is also
possible at the pools, storage systems, hypervisors, and servers levels where the volumes
are analyzed only for the selected resource.
Before starting this use case, we set the tier level of storage pools in our environment. We
have four storage pools and we set three tiers. Figure 7-44 shows the tier levels of the pools
in our use case.
Figure 7-44 Tiering
210
IBM SmartCloud Virtual Storage Center
When the tiers are defined, select the volumes that you want to analyze. We select four
volumes of a test server from the volumes pane. The Analyze Tiering function is available
from either the Actions menu or the context menu, as shown in Figure 7-45. Select Analyze
Tiering.
Figure 7-45 Analyze Tiering
Because we are running a workload on the selected volumes, you can see the performance
monitor for all volumes (Figure 7-46).
Figure 7-46 Volume performance
In this use case, we use I/O rate to perform Analyze Tiering of the volumes. The source
storage pools that are related to the selected volumes are analyzed to determine whether
they meet the workload requirements of the volumes. If the workload requirements of the
volume in its current tier are not met, the volume is a candidate for relocation. Figure 7-47
shows I/O rates for the selected volumes that are used for Analyze Tiering.
Figure 7-47 Volumes I/O rate
Chapter 7. Storage optimization
211
After you select Analyze Tiering, target pools must be selected to optimize the placement of
the volumes (Figure 7-48). We select all tiers because we want to optimize the placement
across all tiers. Click Next.
Figure 7-48 Target pools in Analyze Tiering
Figure 7-49 on page 213 shows Analyze Tiering definitions where you specify thresholds that
are used in the analysis.
First, enter the number of days to use for analysis. By entering the number of days, you are
manually updating the pane so the values for activity are updated. We use 15 days for this
analysis.
If the volumes in the source pool that are assigned to the same hypervisor or server are
assigned to different host connections, the collocation of the volumes is affected. In such
cases, if volumes require optimization, the volumes that are assigned to the same host
connection are kept together. In this case, we do not use collocation because it is not
important for this test server.
You must decide if you want to use the default or the advanced options in Analyze Tiering. In
this case, we are showing you the advanced options so we select Show advanced options,
You can then set up volume thresholds for I/O rate, I/O density, and file usage. If you use the
default options instead, you can set up the Activity limit only for tiers that are analyzed.
The default values for I/O density or I/O rate are not based on any analysis and you need to
change those numbers based on your system requirements. If you do not know what
threshold values to use, a better approach is to consider using the advanced option. The
advanced option implies that you understand the workload of the tiers and that you need this
flexibility.
212
IBM SmartCloud Virtual Storage Center
Figure 7-49 Analyze Tiering definition
In the definition, you set minimum thresholds for each tier of storage. In this case, we have
volumes from three tiers that are analyzed, so thresholds must be set for all 3 tiers.
We specify minimum thresholds for I/O rate for each tier. You can also specify minimum
thresholds for I/O density and file usage or you can combine minimum thresholds for I/O
density and file usage, and for I/O rate and file usage. The workload of the volumes is
evaluated to determine whether the volumes require re-tiering. In this case, we define the
following I/O rates:
򐂰 Tier 1 = 15000 I/O per second
򐂰 Tier 2 = 2000 I/O per second
򐂰 Tier 3 = 500 I/O per second
By setting minimum I/O rate thresholds, volumes are evaluated. For example if the volume in
Tier 2 has more than 15000 I/O per second, it is candidate to move up to Tier 1.
Note: When performing Analyze Tiering at the volume level, the pool activity values are
used with the I/O rate that you set. In this example having an I/O rate higher than 15000 I/O
per second will recommend a move only if the pool activity limit did not exceed it.
Chapter 7. Storage optimization
213
The activity limit must be set because it is used together with I/O rate to Analyze Tiering. You
can calculate the activity limit by clicking the Help me calculate link. This value is set to the
sum of the maximum read I/O rate and the write I/O rate operations that you want to specify
as limits, divided by the average capacity of all of the pools on the same tier. In addition, the
maximum write I/O rate is the highest value that a pool in the storage tier can accommodate
in conjunction with the read I/O and vice versa. They must stay within desired performance
bounds.
An example of calculating the activity limit is shown in Figure 7-50.
Figure 7-50 Calculate Activity Limit
After you enter all the parameters in the definition for Analyze Tiering, click Analyze (see
Figure 7-49 on page 213).
The completed analysis, with recommendations, is shown in Figure 7-51.
Figure 7-51 Analyze Tiering recommendations
According to the analysis, three volumes will be moved to a new tier. Volumes TPC_SVC1_0001
and TPC_SVC1_0002 will be moved down from Tier 1 to Tier 2, and volume TPC_SVC1_2000 will
be moved up from Tier 3 to Tier 2.
214
IBM SmartCloud Virtual Storage Center
Figure 7-52 shows the complete log of the analysis.
Figure 7-52 Analyze Tiering log
You can schedule the task or start it immediately. In this use case, we start it immediately.
Figure 7-53 shows that task is running.
Figure 7-53 Analyze Tiering task running
When the task starts, storage virtualizer uses VDisk copy to move the volumes. The VDisk
copy task creates a copy of the volumes that are moving to a new tier and starts the
synchronization process.
Chapter 7. Storage optimization
215
Figure 7-54 shows the VDisk copy task in the SAN Volume Controller GUI. The new copies of
the analyzed volumes are configured in a new pool, which is in different tier than the original
pool.
Figure 7-54 SAN Volume Controller VDisk copy
216
IBM SmartCloud Virtual Storage Center
The synchronization process updates the copy until it is fully synchronized. During the
synchronization, the workload runs on the test server by using the IO meter tool (Figure 7-55)
and in the SAN Volume Controller GUI (Figure 7-56).
Figure 7-55 IO meter workload
Figure 7-56 SAN Volume Controller GUI: workload performance
Chapter 7. Storage optimization
217
When the copy is fully synchronized, the storage virtualizer splits off a synchronized volume
copy to a new volume; the workload continues. The original volumes are deleted and the
Analyze Tiering task is complete (Figure 7-57).
Figure 7-57 Analyze Tiering task is completed
Figure 7-58 shows the log of the Analyze Tiering task.
Figure 7-58 Analyze Tiering log
218
IBM SmartCloud Virtual Storage Center
The optimized volumes are shown in Figure 7-59. You see that they are moved to a new pool
in a new tier.
Figure 7-59 Volumes after Analyze Tiering task is done
7.7 Use case: Volume transformation
You can use the volume transformation function to convert fully allocated volumes to
thin-provisioned volumes in storage pools. Other volume transformation options (migrate
pool, compression, Easy Tier) use the same wizard as shown in this use case.
Thin provisioning enables the storage systems to present the required capacity to the host
while allocating only the actual used capacity in terms of space on the physical storage
media. By using volume transformation to transform fully allocated volumes to
thin-provisioned volumes, your storage environment is better used, and unused space is
released. This use case shows how we transform a fully allocated volume of 10 GB to a
thin-provisioned volume. We also run a workload that is using approximately 6 GB of data
on that volume.
You can start the volume transformation function from any list of volumes. In this case, we
start the volume transformation function from the Servers panel, and we use our test server
IICTPCRT1-SEC (Figure 7-60). The server has four volumes that are fully allocated.
Figure 7-60 Volumes pane
Chapter 7. Storage optimization
219
We want to transform volume TPC_SVC1_0001 to a thin-provisioned volume. To transform
it, either select Actions → Transform Storage or right-click the volume and select the
Transform Storage option (Figure 7-61).
Figure 7-61 Transform Storage option
Because the pools where the volume is defined are in a capacity pool, we receive a warning
message (Figure 7-62). Click Next.
Figure 7-62 Transform Storage warning
220
IBM SmartCloud Virtual Storage Center
In the next pane (Figure 7-63), select Thin Provisioning and set the option to On. Click Next.
Figure 7-63 Thin Provisioning option
The Transform Storage pane opens (Figure 7-64). Enter the information in the empty fields,
setting the values as required.
Figure 7-64 Thin provisioning details
If you want to transform more volumes, see the Maximum volume capacity field, which
refers to the smallest volume size. This field helps you track how many volumes and the
capacity you have used so far. You can then determine if you want to transform more
volumes.
Chapter 7. Storage optimization
221
Figure 7-65 shows example of multiple volume selection for volume transformation.
Figure 7-65 Multiple volume selection
You can change the default values for thin-provisioned volume according to your
requirements and environment. In this case, we use default values.
The transform volume configuration fields for thin provisioning and their use are as follows:
򐂰 Allocated space
In this field, we enter the real capacity value. Use the drop-down menu to indicate
percentage (%) or an absolute value (expressed in MiB, GiB, TiB).
– Percentage values: In this case the real capacity equals the indicated percentage value
of the current capacity.
– Absolute values: As an example, an absolute value of 1 GiB is the real capacity for the
selected volume.
򐂰 Auto expand
When this option is set to Yes, space is automatically added to the thin-provisioned
volume when needed. The amount of space that is added is determined by the value that
you enter in the Allocated space field.
If Auto expand is set to No, the thin-provisioned volume is taken offline when it runs out of
available space.
򐂰 Warning level
A warning event is generated on the storage system when the percentage of used space
exceeds the value that you specify. You can define a warning level either by specifying a
value in this field, or by moving the warning level slider.
To be able to set a warning level, you must first set Auto expand to Yes.
򐂰 FlashCopy volumes
If the volume you are transforming is a source or a target FlashCopy volume, select Yes.
In the field that opens, specify a grain size, which must match the grain size that is
specified in FlashCopy relationship.
Click Recommend. Tivoli Storage Productivity Center then prepares the transformation plan.
222
IBM SmartCloud Virtual Storage Center
When it completes the Transform Plan pane opens (Figure 7-66).
Figure 7-66 Transform Plan
You can schedule the transform plan or you can start it immediately. In this case, we start it
immediately.
Also in this case, we run a workload on the volume that will be transformed (Figure 7-67). By
running the workload, we show that this task is not disruptive to your server.
Figure 7-67 Volume performance monitor
Chapter 7. Storage optimization
223
When the task starts, the status changes to Running (Figure 7-68).
Figure 7-68 Transform volumes task is running
Also when the task starts, the storage virtualizer uses VDisk copy to transform the volume.
The VDisk copy task creates thin-provisioned copy of a selected fully allocated volume and
starts the synchronization process.
Figure 7-69 shows the VDisk copy task in the SAN Volume Controller GUI and shows that the
new copy of the selected volume is configured.
Figure 7-69 VDisk copy
The synchronization process updates the copy until it is fully synchronized. When the copy is
fully synchronized, the storage virtualizer splits off a synchronized volume copy to a new
thin-provisioned volume and workload will continue on the new thin provisioned volume. The
original fully allocated volume will be deleted after the transform volumes function is done.
Figure 7-70 shows the new thin-provisioned volume.
Figure 7-70 Thin-provisioned volume
The physical allocation on the thin-provisioned volume is 59%, which means that the
allocated capacity is only 5.94 GB. By transforming a fully allocated volume of 10 GB to a
thin-provisioned volume, 4.06 GB is released.
224
IBM SmartCloud Virtual Storage Center
The complete log of the volume transformation function is shown in Figure 7-71.
Figure 7-71 Volume transformation log
Chapter 7. Storage optimization
225
226
IBM SmartCloud Virtual Storage Center
A
Appendix A.
Tivoli Storage FlashCopy
Manager and DB2 scripts
The appendix describes the Tivoli Storage FlashCopy Manager and DB2 scripts pre-freeze
and post-thaw. The scripts also document the steps to create customer scripts before running
a backup. These scripts are written for a Windows and VMware environment. They will reside
in the virtual machine where our customer application is running.
Modifying scripts: These scripts are examples only and must be modified for your
environment.
© Copyright IBM Corp. 2015. All rights reserved.
227
A.1 Scripts
The freeze script puts the DB2 database into a suspend write mode before VMware takes a
snapshot of the virtual machine. This script passes the suspend write operation to the main
executing script. The name of the script is freeze.cmd and it is in the C:\Windows directory.
The last script that is called is the thaw script. This script calls the main execute script and
passes it the resume write command, which takes the database out of a suspend mode and
resumes normal operations. This script is in the C:\Windows directory.
Note: These scripts are examples only and will need to be modified for each user
environment.
Example A-1 shows the DB2 pre-freeze (freeze) and post-thaw (thaw) scripts, and also an
example of the wrapper script needed to call the main script. The main script, also shown, is
called by the freeze and thaw scripts.
Example A-1 DB2 pre-freeze and post-thaw
The script below is a wrapper script that calls the main scripts depending on the
operation, freeze or thaw. The name of the script is vcb.bat and is located in the
following directory “C:\Program Files\VMware\VMware Tools\backupScripts.d”
Wrapper script:
echo off
if "%1" == "freeze" goto doFreeze
goto doThaw
:doFreeze
call c:\windows\freeze.cmd
goto EOF
:doThaw
call c:\windows\thaw.cmd
:EOF
Freeze script:
The script below is the freeze script, this script puts the DB2 database into suspend write
mode prior to VMware taking a snapshot of the virtual machine. This script passes the
suspend write operation to the main executing script. The name of the script is freeze.cmd
and is located in the C:\Windows directory you can enable logging and set the log directory
in this script
@echo off
@setlocal
@rem
@rem
@rem
@rem
@rem
@rem
@rem
@rem
@rem
@rem
@rem
@rem
@rem
@rem
228
**************************************************************************
db2_preconpoint.cmd
Description: Pre consistency point script for DB2 backup.
Functionality:
- Connects to the local database to back up.
- Calls a SQL script that suspends write operations for the database to be backed up.
- Terminates database connection.
Authorization:
You must run the script as an user id with sysadm, sysctrl, or sysmaint authority.
Exit with Return Code:
IBM SmartCloud Virtual Storage Center
@rem 0, if execution is successful.
@rem other, if error encountered.
@rem
@rem **************************************************************************
@rem version 0.6
@rem last modification 20080806
set SCRIPT_LEVEL=20080806
@rem author
@rem **************************************************************************
@rem **************************************************************************
@rem !!! ADJUST THE FOLLOWING VARIABLES IF YOU WISH TO USE OTHER THAN THE DEFAULT VALUES
!!!
@rem doLogging - Flag to create a trace log; possible values: TRUE, FALSE
@rem logDir
- Trace log directory
set doLogging=TRUE
set logDir=G:\logs
@rem !!! END OF SETTING VARIABLES !!!
@rem !!! DO NOT MODIFY BELOW THIS LINE OF CODE !!!
@rem **************************************************************************
@rem Create the log dir if there is not yet.
dir "%logDir%" 1>NUL 2>&1
set /A RC=%errorlevel%
if "%RC%" equ "0" goto isLogDir
mkdir "%logDir%"
set /A RC=%errorlevel%
if "%RC%" equ "0" goto isLogDir
@echo Error: Cannot create the log directory "%logDir%"
goto byebye2
:isLogDir
set /A RC=0
if /i "%doLogging%" NEQ "true" goto noTrace
@rem traceLogPrefix
- prefix for the log file name
set traceLogPrefix=%~n0_
@rem traceLogExtension - log file extension
set traceLogExtension=log
@rem Set the file name of the trace log to contain the current date and time.
goto parsDate
:datePrsd
goto parsTime
:timePrsd
set
traceLog="%logDir%\%traceLogPrefix%%_year%%_month%%_day%%_hh%%_min%%_ss%%_ds%.%traceLogExte
nsion%"
goto commence
:noTrace
set traceLog=NUL
:commence
@echo %date% %time% Running %~0 ...>>%traceLog%
@echo %date% %time% Script level %SCRIPT_LEVEL% >>%traceLog%
@rem Delete db2.err file if there is any.
set db2err=%logDir%\db2.err
if not exist "%db2err%" goto runCLP
@echo %date% %time% del /F /Q /A:H "%db2err%" 1>>%traceLog%
del /F /Q /A:H "%db2err%" 1>>%traceLog% 2>&1
Appendix A. Tivoli Storage FlashCopy Manager and DB2 scripts
229
:runCLP
@rem Run the DB2 CLP command.
set CWD=%~dp0
set execSQL="%CWD%db2_executeSQL.cmd"
@echo %date% %time% call db2cmd -w -c -i """%execSQL%""" SUSPEND """%traceLog%"""
1>>%traceLog%
call db2cmd -w -c -i """%execSQL%""" SUSPEND """%traceLog%"""
set /A RC=%errorlevel%
@echo %date% %time% executeSQL.cmd exited with RC=%RC% 1>>%traceLog%
goto byebye
@rem **************************************************************************
@rem Parse generic system date to get the numerical values of current month, day, and
year.
@rem Set variables _month, _day, _year
@rem ********************************************************************
:parsDate
set $tok=1-3
for /f "tokens=1 delims=.:/-, " %%u in ('date /t') do set $d1=%%u
if "%$d1:~0,1%" GTR "9" set $tok=2-4
for /f "tokens=%$tok% delims=.:/-, " %%u in ('date /t') do (
for /f "skip=1 tokens=2-4 delims=/-,()." %%x in ('echo.^|date') do (
set %%x=%%u
set %%y=%%v
set %%z=%%w
set _month=%%u
set _day=%%v
set _year=%%w
set $d1=
set $tok=))
goto datePrsd
@rem ********************************************************************
@rem Parse generic system time to get the numerical values of current hour, minute,
second, and second division.
@rem Set variables _hh, _min, _ss, _ds with the time elements as 2 digit values (24 hr
clock.)
@rem ********************************************************************
:parsTime
for /f "tokens=5-8 delims=:., " %%a in ('echo:^|time') do (
set _hh=%%a
set _min=%%b
set _ss=%%c
set _ds=%%d
)
if %_hh% LSS 10 set _hh=0%_hh%
goto timePrsd
:byebye
@echo %date% %time% Exit %~0 >>%traceLog%
@echo %date% %time% RC=%RC% >>%traceLog%
:byebye2
@endlocal& exit /B %RC%
Main script:
The script below is the main executing script. This is called by the freeze and thaw
scripts and passes either the suspend write or resume write depending on the operation. The
230
IBM SmartCloud Virtual Storage Center
name of it is db2_executeSQL.cmd and is located in C:\Windows. In this script you need to
define the following parameters:
򐂰 dbuser
򐂰 dbpass
򐂰 dbinst
򐂰 dbname
@echo off
@setlocal
@rem **************************************************************************
@rem executeSQL.cmd
@rem
@rem Script used by TSM FastBack for backing up DB2.
@rem It suspends and resumes write I/O for a database.
@rem It also can restart the database with write resume option.
@rem
@rem Usage: executeSQL.cmd <operation> <trace-log>
@rem
@rem All parameters are required:
@rem operation - Operation to be run against a database
@rem
Possible values (case insensitive) are:
@rem
SUSPEND, to temporarily suspend database write I/O
@rem
RESUME, to resume write I/O for the database
@rem
RESTART, to restart the database with WRITE RESUME
@rem trace-log - Absolute path of the trace log
@rem
@rem IMPORTANT: You also need to set variables for db user, db password, db name, db
instance.
@rem
@rem Exit codes:
@rem
0, if execution is successful
@rem -1, if there are missing parameter(s).
@rem -99, if an attempt to connect to the database returned an error message,
@rem
indicating that the database must be restarted.
@rem other value, if other error types encountered.
@rem **************************************************************************
@rem
@rem **************************************************************************
@rem version 0.5
@rem last modification 20080717
@rem author
@rem **************************************************************************
@rem **************************************************************************
@rem !!! ADJUST THE FOLLOWING VARIABLES IF YOU WISH TO USE OTHER THAN THE DEFAULT VALUES
!!!
@rem dbuser
- User ID having with sysadm, sysctrl, or sysmaint authority.
@rem dbpass
- Password of that user ID
@rem dbinst
- DB2 instance name that runs the database
@rem dbname
- Database alias to back up
set dbuser=db2admin
set dbpass=Object00
set dbinst=DB2
set dbname=TPCDB
@rem !!! END OF SETTING VARIABLES !!!
@rem !!! DO NOT MODIFY BELOW THIS LINE OF CODE !!!
@rem **************************************************************************
if not ".%~2" equ "." goto SETENV
Appendix A. Tivoli Storage FlashCopy Manager and DB2 scripts
231
@echo Error: Missing parameter(s).
@echo Usage: executeSQL.cmd ^<operation^>
set /A RC=-1
goto byebye
^<trace-log^>
:SETENV
set operation=%1
set log=%2
@rem Check for empty variables
if "%dbuser%" EQU "" set emptyVar=dbuser
if "%dbpass%" EQU "" set emptyVar=dbpass
if "%dbinst%" EQU "" set emptyVar=dbinst
if "%dbname%" EQU "" set emptyVar=dbname
if defined emptyVar goto invalid
@echo %date% %time% Running %~0...>>%log%
@echo
dbuser
: %dbuser% >>%log%
@echo
dbpass
: ***** >>%log%
@echo
dbinst
: %dbinst% >>%log%
@echo
dbname
: %dbname% >>%log%
@echo
operation
: %operation% >>%log%
@echo
log
: %log% >>%log%
@rem **************************************************************************
@rem ERROR_CONNECTION is the return code in case of connection hanging.
set /A ERROR_CONNECTION=-99
set CLP_FLAGS=-es -ov
@echo %date% %time% set DB2INSTANCE=%dbinst% >>%log%
set DB2INSTANCE=%dbinst%
if /i "%operation%" equ "RESTART" goto restart
@rem **************************************************************************
@rem Create a file called db2.err in %logDir% directory.
@rem Write %ERROR_CONNECTION% in it.
@rem This should prevent the situation in which the subsequent CONNECT command
@rem hangs for any reason when attempting to resume the database.
@rem The script is expected to get cancelled by TSM FastBack as a result of time-out
period expires.
@rem Subsequently this assumes that post sbapshot script will restart the database.
if /i "%operation%" neq "RESUME" goto connect
set /A RC=%ERROR_CONNECTION%
set BAILOUT=FALSE
goto noteRC
@rem **************************************************************************
:connect
set /A RC=0
set BAILOUT=TRUE
@echo %date% %time% Connecting to %dbname% database... 1>>%log%
db2 %CLP_FLAGS% CONNECT TO %dbname% USER %dbuser% USING %dbpass% 1>>%log% 2>&1
set /A RC=%errorlevel%
@echo.>>%log%
if "%RC%" equ "0" goto connOK
@echo %date% %time% Error: Connection to %dbname% database failed RC=%RC%.>>%log%
@rem ADM6024C The database cannot be restarted because table spaces cannot be brought
ONLINE
@rem
as a result of an outstanding WRITE SUSPEND.
232
IBM SmartCloud Virtual Storage Center
@rem
To restart the database specify WRITE RESUME on the RESTART DATABASE
command.
::findstr /r "ADM6024C" %log%
::if %errorlevel% equ 0 set /A RC=%ERROR_CONNECTION%
if /i "%operation%" equ "RESUME" set /A RC=%ERROR_CONNECTION%
goto noteRC
:connOK
@echo %date% %time% Successfully connected to %dbname% database.>>%log%
@rem Now run the DB2 command.
if /i "%operation%" equ "SUSPEND" @echo %date% %time% Temporarily suspend writing to
database... 1>>%log%
if /i "%operation%" equ "RESUME" @echo %date% %time% Resume writing to database... 1>>%log%
db2 %CLP_FLAGS% SET WRITE %operation% FOR DATABASE 1>>%log% 2>&1
set /A RC=%errorlevel%
@echo.>>%log%
if "%RC%" neq "0" @echo %date% %time% Error: Command execution failed RC=%RC% 1>>%log%
goto terminat
:restart
@echo %date% %time% Restart %dbname% database... 1>>%log%
db2 %CLP_FLAGS% RESTART DATABASE %dbname% USER %dbuser% USING %dbpass% WRITE RESUME
1>>%log% 2>&1
set /A RC=%errorlevel%
@echo.>>%log%
if "%RC%" neq "0" @echo %date% %time% Error: Restart %dbname% database failed RC=%RC%
1>>%log%
goto terminat
:terminat
@echo %date% %time% Terminate the CLP's back-end process... 1>>%log%
db2 %CLP_FLAGS% TERMINATE 1>>%log% 2>&1
set /A RC2=%errorlevel%
@echo.>>%log%
if "%RC2%" neq "0" @echo %date% %time% Error: Failed to terminate the CLP's back-end
process RC=%RC2% 1>>%log%
if "%RC%" equ "0" set /A RC=%RC2%
set BAILOUT=TRUE
if /i "%operation%" equ "RESTART" goto byebye
goto noteRC
:noteRC
@rem Write RC to db2.err file
set CWD=%~dp2
set db2err=%CWD%db2.err
if not exist "%db2err%" goto writeRC
@echo %date% %time% del /F /Q /A:H "%db2err%" 1>>%log%
del /F /Q /A:H "%db2err%" 1>>%log% 2>&1
:writeRC
@echo %RC% 1>"%db2err%"
@echo %date% %time% type "%db2err%">>%log%
type "%db2err%" 1>>%log%
@echo %date% %time% attrib +H "%db2err%" 1>>%log%
attrib +H "%db2err%" 1>>%log% 2>&1
if /i "%BAILOUT%" equ "TRUE" goto byebye
@rem for pre snapshot script
goto connect
:invalid
@echo %date% %time% Error: Variable %emptyVar% cannot be set to empty value.>>%log%
Appendix A. Tivoli Storage FlashCopy Manager and DB2 scripts
233
set /A RC=-1
goto byebye
:byebye
@endlocal& exit /B %RC%
Thaw script:
The last script that is called is the thaw script. It calls the main execute script and
passes it the resume write command which take the database out of suspend mode and resumes
normal operations. This script is located in the C:\Windows directory and is called once
the vmware snapshot creation has completed.
@echo off
@setlocal
@rem **************************************************************************
@rem db2_presnapshot.cmd
@rem
@rem Description: Pre snapshot script for DB2 backup.
@rem
@rem Functionality:
@rem - Connects to the local database to back up.
@rem - Calls a SQL script that resumes write operations for the database.
@rem - Terminates database connection.
@rem
@rem Authorization:
@rem You must run the script as an user id with sysadm, sysctrl, or sysmaint authority.
@rem
@rem Exit with Return Code:
@rem 0, if execution is successful.
@rem other value, if error encountered.
@rem
@rem **************************************************************************
@rem version 0.6
@rem last modification 20080806
set SCRIPT_LEVEL=20080806
@rem author
@rem **************************************************************************
@rem **************************************************************************
@rem !!! ADJUST THE FOLLOWING VARIABLES IF YOU WISH TO USE OTHER THAN THE DEFAULT VALUES
!!!
@rem doLogging - Flag to create a trace log; possible values: TRUE, FALSE
@rem logDir
- Trace log directory
set doLogging=TRUE
set logDir=G:\logs
@rem !!! END OF SETTING VARIABLES !!!
@rem !!! DO NOT MODIFY BELOW THIS LINE OF CODE !!!
@rem **************************************************************************
@rem Create the log dir if there is not yet.
dir "%logDir%" 1>NUL 2>&1
set /A RC=%errorlevel%
if "%RC%" equ "0" goto isLogDir
mkdir "%logDir%"
set /A RC=%errorlevel%
if "%RC%" equ "0" goto isLogDir
@echo Error: Cannot create the log directory "%logDir%"
goto byebye2
234
IBM SmartCloud Virtual Storage Center
:isLogDir
set /A RC=0
set /A ERROR_CONNECTION=-99
if /i "%doLogging%" NEQ "true" goto noTrace
@rem traceLogPrefix
- prefix for the log file name
set traceLogPrefix=%~n0_
@rem traceLogExtension - log file extension
set traceLogExtension=log
@rem Set the file name of the trace log to contain date and time
goto parsDate
:datePrsd
goto parsTime
:timePrsd
set
traceLog="%logDir%\%traceLogPrefix%%_year%%_month%%_day%%_hh%%_min%%_ss%%_ds%.%traceLogExte
nsion%"
goto commence
:noTrace
set traceLog=NUL
:commence
@echo %date% %time% Running %~0 ...>>%traceLog%
@echo %date% %time% Script level %SCRIPT_LEVEL% >>%traceLog%
@rem Run the DB2 CLP command.
set CWD=%~dp0
set execSQL="%CWD%db2_executeSQL.cmd"
@echo %date% %time% call db2cmd -w -c -i """%execSQL%""" RESUME """%traceLog%"""
1>>%traceLog%
call db2cmd -w -c -i """%execSQL%""" RESUME """%traceLog%"""
set /A RC=%errorlevel%
@echo %date% %time% executeSQL.cmd exited with RC=%RC% 1>>%traceLog%
goto byebye
@rem **************************************************************************
@rem Parse generic system date to get the numerical values of current month, day, and
year.
@rem Set variables _month, _day, _year
@rem ********************************************************************
:parsDate
set $tok=1-3
for /f "tokens=1 delims=.:/-, " %%u in ('date /t') do set $d1=%%u
if "%$d1:~0,1%" GTR "9" set $tok=2-4
for /f "tokens=%$tok% delims=.:/-, " %%u in ('date /t') do (
for /f "skip=1 tokens=2-4 delims=/-,()." %%x in ('echo.^|date') do (
set %%x=%%u
set %%y=%%v
set %%z=%%w
set _month=%%u
set _day=%%v
set _year=%%w
set $d1=
set $tok=))
goto datePrsd
@rem ********************************************************************
Appendix A. Tivoli Storage FlashCopy Manager and DB2 scripts
235
@rem Parse generic system time to get the numerical values of current hour, minute,
second, and second division.
@rem Set variables _hh, _min, _ss, _ds with the time elements as 2 digit values (24 hr
clock.)
@rem ********************************************************************
:parsTime
for /f "tokens=5-8 delims=:., " %%a in ('echo:^|time') do (
set _hh=%%a
set _min=%%b
set _ss=%%c
set _ds=%%d
)
if %_hh% LSS 10 set _hh=0%_hh%
goto timePrsd
:byebye
@echo %date% %time% Exit %~0 >>%traceLog%
@echo %date% %time% RC=%RC% >>%traceLog%
:byebye2
@endlocal& exit /B %RC%
236
IBM SmartCloud Virtual Storage Center
Related publications
The publications listed in this section are considered particularly suitable for a more detailed
discussion of the topics covered in this book.
IBM Redbooks
The following IBM Redbooks publications provide additional information about the topic in this
document. Some publications referenced in this list might be available in softcopy only.
򐂰 IBM SAN and SVC Stretched Cluster and VMware Solution Implementation, SG24-8072
򐂰 IBM SAN Volume Controller Enhanced Stretched Cluster with VMware, SG24-8211
򐂰 IBM Tivoli Storage Productivity Center: Beyond the Basics, SG24-8236
򐂰 IBM Tivoli Storage Productivity Center V5.2 Release Guide, SG24-8204
򐂰 Implementing the IBM Storwize V7000 Gen2, SG24-8244
򐂰 Implementing the IBM System Storage SAN Volume Controller V7.4, SG24-7933
򐂰 Optimize your investments with IBM Tivoli Storage Productivity Center and IBM
SmartCloud Virtual Storage Center, TIPS1210
You can search for, view, download or order these documents and other Redbooks,
Redpapers, Web Docs, draft and additional materials, at the following website:
ibm.com/redbooks
Other publications
These publications are also relevant as further information sources:
򐂰 IBM Tivoli Storage Productivity Center, Version 5.2.4 User's Guide, SC27-4060
򐂰 IBM Tivoli Storage Productivity Center, Version 5.2.4, Administrator's Guide, SC27-4859
Online resources
These websites are also relevant as further information sources:
򐂰 IBM SmartCloud Virtual Storage Center Wiki
http://ibm.co/H0Ckqb
򐂰 Tivoli Storage Productivity Center welcome page (at the IBM Knowledge Center):
http://www.ibm.com/support/knowledgecenter/SSNE44/welcome?lang=en
򐂰 IBM SmartCloud Virtual Storage Center Storage Analytics Engine, Version 5.2.4, Quick
Start Guide
http://www.ibm.com/support/knowledgecenter/api/content/nl/en-us/SSNE44_5.2.4/co
m.ibm.tpc_V524.doc/qstrtvsc.pdf
© Copyright IBM Corp. 2015. All rights reserved.
237
Help from IBM
IBM Support and downloads
ibm.com/support
IBM Global Services
ibm.com/services
238
IBM SmartCloud Virtual Storage Center
IBM SmartCloud Virtual Storage
Center
IBM SmartCloud Virtual Storage
Center
IBM SmartCloud Virtual Storage Center
IBM SmartCloud Virtual Storage Center
(0.2”spine)
0.17”<->0.473”
90<->249 pages
IBM SmartCloud Virtual Storage
Center
IBM SmartCloud Virtual Storage
Center
Back cover
®
IBM SmartCloud
Virtual Storage Center
®
Explore virtualization,
storage, and cloud for
your infrastructure
IBM SmartCloud Virtual Storage Center provides efficient virtualization
and management of heterogeneous storage systems. It facilitates
migration to an agile cloud architecture that can optimize storage
availability and performance, while helping to reduce costs.
Accelerate business
insights with
advanced analytics
IBM SmartCloud Virtual Storage Center (VSC) helps convert existing
storage to IBM Smarter Storage, providing more room for data growth
and simplified storage administration.
Follow along with
sample guided use
cases
This IBM Redbooks publication gives an overview of the concepts of
software-defined environment (SDE) and software-defined storage
(SDS), and how they work together with VSC. It explores the
architecture, components, and interfaces, providing details of VSC and
how to use it.
It also includes practical scenarios and use cases, helpful for client VSC
business environments, with a focus on the following topics:
򐂰
򐂰
򐂰
򐂰
򐂰
򐂰
򐂰
Introductory concepts
VSC components and available integrations
Storage management component of VSC
Storage virtualization component of VSC
Application aware data protection component of VSC
VSC storage provisioning
VSC storage optimization
This book is primarily for storage administrators, users who are
responsible for maintaining IT and business infrastructures, and
anyone who wants to learn more about IBM SmartCloud Virtual
Storage Center.
INTERNATIONAL
TECHNICAL
SUPPORT
ORGANIZATION
BUILDING TECHNICAL
INFORMATION BASED ON
PRACTICAL EXPERIENCE
IBM Redbooks are developed
by the IBM International
Technical Support
Organization. Experts from
IBM, Customers and Partners
from around the world create
timely technical information
based on realistic scenarios.
Specific recommendations
are provided to help you
implement IT solutions more
effectively in your
environment.
For more information:
ibm.com/redbooks
SG24-8239-00
ISBN 0738440434
Was this manual useful for you? yes no
Thank you for your participation!

* Your assessment is very important for improving the work of artificial intelligence, which forms the content of this project

Download PDF

advertising