REGISTER TODAY!http://ibmsystemsmag.webex.comB Click on “Event Center”
DATE & TIMEWednesday, March 9, 2011 – 11 a.m. EST
WHO SHOULD ATTENDVP/Director for IT, development managers, project managers, architects, analysts and systems programmers, developers, deployment engineers
AGENDAPlease attend a complimentary one-hour webcast to learn how the new and enhanced development tools from IBM Rational® can help simplify development and delivery of multiplatform applications that span z/OS®, Linux on System z®, AIX® on POWER®, Linux on POWER(R), and Linux on System x® under a single management domain. The Rational software development platform offers developers an integrated set of tools, best practices and workbenches to overcome cross-platform development challenges, including capabilities to revitalize applications, empower developers, unify teams and exploit the new zEnterprise™ infrastructure. These elements all contribute to help simplify and accelerate the creation, maintenance and enhancement of enterprise applications which run on multiple platforms across the enterprise. These are some of the topics we’ll cover during this webcast: B How Rational solutions work with zEnterprise systems to enable mainframe development on a Linux- based PC or on System z B A platform which allows the entire development team to stay informed of development, build, and test status as updates are made across the development project B Tools to enable developers to discover and analyze existing application assets in order to lower the cost of reusing or lower the risk of enhancing, or use as components or services for new applications
SPECIAL OFFERRegister to receive an IBM System z “Can do-it eKit” and Enterprise Modernization Sandbox for System z.
FEATURINGB Peter S. Wassel, Featured Speaker Program Director, System z and Cross-Platform Segments, IBM Rational Software
B Doug Rock, Moderator Publisher, IBM Systems Magazine
LEARN MORE about Modern Application Development and Deployment at IBM’s System z Technology Summit, coming soon to a city near you!
www.ibm.com/software/os/systemz/summit/track3.html
WEBINAR | Register now!
WHY ACCELERATE ENTERPRISE MODERNIZATION WITH IBM RATIONAL DESIGN AND DEVELOPMENT SOLUTIONS?A System z Developer’s Journey Through the Application Lifecycle
B Tim Hahn, Featured Speaker IBM Distinguished Engineer, IBM Master Inventor, IBM Rational Software
Address 8 Service-Delivery Challenges Page 40
March/April 2011 ibmsystemsmag.com
MAINFRAME
Cognos on System z Reduces BI CostsPage 36
Companies stay ahead of the competition using IBM's end-to-end offerings Page 32
>aZZUZS[Z�
-ZMXe`UO_
© Dan Cardiff - iStockphoto
Not everyone trains to be
a Search and Rescue Swimmer...
z / V M P E R F O R M A N C E T O O L S F R O M z / V M P E R F O R M A N C E P E O P L E
...and even fewer make the grade.
Search and Rescue is often done during the worst conditions imaginable. It’s a tough job and one that demands the best people and tools you can fi nd. The same can be said for those managing complex z/VM environments. Migrating Linux servers to z/VM creates many challenges and fi nding the right tools and training is nearly impossible. It’s why Velocity Software continues to offer both free education and some of the best z/VM tools in the business. In 2010 we trained more than 100 customers, business partners, and IBMers at no charge. We intend to expand our topics and more than double that number during 2011.
But training is just one factor when it comes to successfully managing your z/VM complex. To simplify the way you manage z/VM and instantly improve productivity, Velocity is announcing a web based option of zVPS (Velocity’s Performance Suite for System z). zPRO (z/VM Professional) provides an intuitive systems management portal for z/VM and Linux on System z. It can be used to clone servers (Linux and others), manage your Directory, handle document retrieval, manage Spool Files, and administer TCP/IP and RACF defi nitions as you develop your cloud environment with IBM System z. zPRO can be installed in minutes along with zVPS. Why make the job any tougher than it is already? Take a little of the “chill” out of being a z/VM professional. Contact us for more information about pricing, demos, and trials.
zVPS from Velocity Software. It’s the smart thing to do.
Offi ces: Mountain View, California; Irving, Texas; Columbus, Ohio; Mannheim, Germanywww.velocitysoftware.com USA: 877.964.8867 International: 1.650.964.8867 or +49 (0) 621.373.844
� !("���$�+����������"���$#�" # "& ��
'� ���������� �
$� �'�(�$� ������"�� �
�#������ ��� " & "�
'�����
������������ �������������
������������
������"������ "%#�
"���$����(�#�!��#�
���"�������������$�
$ ����!������( %"�!� !���
� "��#%���##%���� �
� "��!" %�$&��'����
���!���( %����$���' "� �
���##��%��$(� ���"&����
������"������ "%#�#���
!�"$� �$���������"����
����#$"�$��(���$�#�� $�����#$�
��"��!�$ ����"����
���������$�"�#! �#��$�#�
�� ���!" %�$&$(�������
�#��� �$ ����!�( %���$�
� "��&��%��" ��( %"�
���"����!��$ "���
��!��"����#$�"��� "��
!" %�$&��
����"#$����������$�
" ����)������"����
�� "%#� "�������$���#��
���������$�*�#��&�������
$ �(��
��$" %����������"������ "%#�" ���������� � ��#���
�
Mainframe MARCH/APRIL 2011
Inside
COVER STORYDriving Outcomes ± 32
Successful companies fuel
business with predictive analytics.
By Kathy Konkel, cover illustration
by Oliver Burston
FEATUREHurdling Barriers
to Success ± 36
IBM demonstrates Cognos BI
on Systems z is less expensive
than x86 deployments.
By Rebecca Wormleighton
CASE STUDYA Sound Investment ± 28
Citigroup tames its backup
environment with dedicated
mainframes. By Jim Utsler
28
IBM has some strong algorithms built into the
machines, as well as crypto engines and certificate management.”
— Martin Kennedy, managing
director of enterprise-systems
infrastructure Citigroup Inc.
2 M A R C H / A P R I L 2 0 1 1 i b m s y s t e m s m a g . c o m / m a i n f r a m e / / m a i n f r a m e24
32
36
Inside
2
DEPARTMENTS
Editor’s Desk ± 12
Predicting a Good Read
Dashboard ± 14
Charging Ahead
Data Display ± 16
Smarter Cities Large and Small
Think Smarter ± 18
Capture Insight From Real-Time
Operational Data With Smart
Analytics Systems
Insider ± 22
Agnostic Development Tools Support
Mainframe-Centric Businesses
Trends ± 26
DataPower Support for zEnterprise
Extends its Multiplatform
Capabilities
Tips and Techniques ± 40
Overcome Eight Key Service-Delivery
Challenges and Reduce Costs
Solutions ± 45
7BOHVBSE�$POGJHVSBUJPO�.BOBHFS�t�
;&/�t�$MFWFS7JFX�GPS�5$1�*1�W������t�
PlexSpy Application Status Monitor
W�S��t�1SFEJDUJWF�3FDBMM�t�/FUVJUJWF���
Resources ± 47
Advertisers’ Index
Stop Run ± 48
Bijan Davari Changed the Computer
Chip
®
Mainframe JULY/AUGUST 2010
M A R C H / A P R I L 2 0 1 1 i b m s y s t e m s m a g . c o m / m a i n f r a m e2
Mainframe MARCH/APRIL 2011
6
14
22
48
26
This is just a sampling of the IBM Systems Magazine content
available at the click of a mouse. Visit us online frequently for
Web-exclusive articles, blog entries, audio tidbits and more.
On the Web
ibmsystemsmag.com
M A R C H / A P R I L 2 0 1 1 i b m s y s t e m s m a g . c o m / m a i n f r a m e8
Weigh in on The BuzzCan U.S. kids overcome the geek stigma? Copy Editor Morgon Mae
Schultz posed the question to readers via our blog, and you weighed
in. See what everyone is “buzzing” about:
http://ibmsystemsmag.blogs.com/the_buzz/2010/11/can-us-kids-
overcome-geek-stigma.html
Only OnlineSee what additional content is popping up in our EXTRA e-newsletter:
“Common Tools, Multiple Platforms,” by Tim Hahn,
Rational enables development in a multiplatform runtime
environments
www.ibmsystemsmag.com/mainframe/
enewsletterexclusive/35616p1.aspx
“Safely Concealed,” by Jim Utsler,
IBM Identity Mixer is poised to change how Web users reveal personal data
www.ibmsystemsmag.com/mainframe/ibm_research_
identitymixer/35550p1.aspx
“The Three Ps of Web Services,” by Jim Schesvold,
Retail companies must address more than performance
www.ibmsystemsmag.com/mainframe/holiday_
performance2/35685p1.aspx
Beyond Flashcards
DB2utor Troy Coleman shares a new
DB2 resource dubbed a “Flashbook.”
http://ibmsystemsmag.blogs.
com/db2utor/2011/01/a-great-db2-10-
resource.html
BACKUP DATA TRAVELS ON FICON CHANNELS…NOT YOUR TCP/IP LINKSFDRSOS…is a high speed, cross platform Open Systems diskbackup and rapid recovery solution…Data Travels on FICONChannels not your TCP/IP Network.
No longer will backups…take too long…congest networks…or leave you worrying about the prospects of a reliable recovery!
� Take distributed data backup off communication networks.
� Eliminate the need for distributed backup servers.
� Ensure backup no longer constrains production.
� Empower Open Systems backup with System z RAS.
� Employ your existing mainframe tape management & security.
FDRSOS… and FlashCopy or TimeFinder, lets you keep critical and revenue generating applications online withoutcompromising information security.
FDRSOS…no other solution has the capability of using high performance Systems z FICON channels to directly readand write the same disk volumes used by the Open Systemsbusiness applications on DS8700 and VMAX storage.
FDRSOS…provides the ultimate in z/OS distributed data protection and disaster recovery for Linux on System z, AIX,Linux x86-64, NetWare, OES2, UNIX and Windows.
FDRSOS… can help you consolidate hundreds of distributedOpen Systems backup servers onto one z/OS backup serverrunning on System z.
For More Information and a No-Obligation FREE Trialcontact: 973-890-7300, [email protected] orvisit: www.innovationdp.fdr.com
FICON FICONCLIENTS
FIBRE
Application Disk on IBM DS8700 and EMC VMAX Storage
IBMzEnterprise
FICONTape/VTS
Backup Data Flow Backup Data Flow
Distributed Data Disaster Recovery Protection
CORPORATE HEADQUARTERS: ����!'/(-.,+��0(���)//*(�'**.����� �����2��������� ��� �2�'1���������� �����E-mail: [email protected] 2�[email protected] 2�http:/ / www.innovationdp.fdr.com
EUROPEAN FRANCE GERMANY NETHERLANDS UNITED KINGDOM NORDIC COUNTRIESOFFICES: ����������� ������� ������� �� ����� ���������� ���������� � ��� �������������������������������
EDITORIAL
EXECUTIVE PUBLISHER
Diane Rowell
PUBLISHER
Doug Rock
EXECUTIVE EDITOR
Evelyn Hoover
MANAGING EDITOR
Natalie Boike
COPY EDITORS
Ann Schroeder
Caroline Vitse
SENIOR WRITER
Jim Utsler
TECHNICAL EDITORS
Ron Higgin
Jim Schesvold
PRODUCTION
ART DIRECTOR
Jill Adler
PRODUCTION MANAGER
Jonathan Benson
PROJECT MANAGER
Kelly Daugherty
ASSOCIATE TECHNICAL
PRODUCER
David Waters
CIRCULATION
CIRCULATION DIRECTOR
Jane Breckinridge
CIRCULATION MANAGER
Linda Holm
CIRCULATION
COORDINATOR
Carin Russell
FULFILLMENT
COORDINATOR
Rebecca Bogema
ADVERTISING/SALES
ASSOCIATE PUBLISHER
Mari Adamson-Bray
(612) 336-9241
SENIOR ACCOUNT
MANAGER, NORTHEAST,
NORTHWEST & CANADA
Kathy Ingulsrud
(612) 313-1785
ACCOUNT EXECUTIVE,
MIDWEST & EUROPE
Darryl Rowell
(612) 313-1781
ACCOUNT EXECUTIVE,
SOUTHEAST, SOUTHWEST
& AUSTRALIA
Lisa Kilwein
(574) 988-0011
SALES AND MARKETING
DEVELOPMENT MANAGER
Katie Swanson
IBM EDITORIAL BOARD
ERIC BAXLEY
Director, System z Software
Marketing
MARIANNE CARTER
Software Product Manager
DB2 for z/OS Program Offerings
PAULO CARVAO
Vice President,
System z Global sales
IBM Systems & Technology Group
DORIS CONTI
System z Strategy Manager
PAUL DIMARZIO
STG System z Brand,
Enterprise Integration Strategy
JOE DORIA
Director, Global Marketing
IBM System z
WILLIE FAVERO
Senior Certified Consulting IT
Software Specialist
Data Warehouse for System z
Swat Team
JUAN GASTELU
Manager, Data Center
Optimization Marketing
IBM Systems and
Technology Group
SIMON HARES
System z Marketing Manager
IBM Sales and Distribution
BOB HOEY
Global Major Markets
General Manager
JUERGEN HOLTZ
Design and Development, Tivoli
System Automation Technologies
KURT JOHNSON
System z Market Management
and Marketing Strategy
GREG LOTKO
Vice president and business-
line executive, System z
ALLEN MARIN
IBM System Storage Enterprise
Disk Marketing
MARY MOORE
STG Business Resilience
Offerings
DOUG PEARSON
Manager, System z Platform
Marketing
JIM PORELL
IBM Distinguished Engineer
System z Business Development
JACQUI (JAX) SHAWLEY
WorldWide Market Manager
LMI Program Marketing –
System z Software
JACK YUAN
IMS TM and Connect
Development
Lead IBM Software Group/
Database Management
Mainframe edition
AIX
DB2
ESCON
FICON
IBM
IBM logo
MVS
OS/390
POWER
S/390
System i
System p
System Storage
System z
System z9
System z10
Tivoli
TotalStorage
VM/ESA
VSE/ESA
WebSphere
System x
z/OS
z/VM
Java and all Java-based trademarks are trademarks of Oracle Corporation, in the United States
and other countries. Microsoft, Windows and Windows NT are trademarks of Microsoft Corp., in
the United States and other countries. Intel, MMX and Pentium are trademarks of Intel Corp. in
the United States and other countries. UNIX is a registered trademark of The Open Group in the
United States and other countries. SET is a trademark owned by SET Secure Electronic Transaction
LLC. Linux is a trademark of Linus Torvalds in the United States, other countries, or both. Linear
Tape-Open, LTO, the LTO logo, Ultrium, and the Ultrium logo are trademarks of Certance, HP and
IBM in the U.S. and other countries. Storwize and the Storwize logo are trademarks or registered
trademarks of Storwize Inc., an IBM Company. Other company, product and service names may be
trademarks or service marks of others.
Articles appearing in IBM Systems Magazine, Mainframe edition may have been published in previ-
ous IBM Systems Magazine editions.
Reprints: To order reprints, contact Kelly Carver (612) 336-9280.
IBM Systems Magazine, Mainframe edition (ISSN# 1544-8819) is published bimonthly by MSP
TechMedia, 220 South Sixth St., Suite 500, Minneapolis, MN 55402. Publications Agreement
No. 40063731, Canadian Return Address, Pitney Bowes, Station A, PO Box 54, Windsor, Ontario
Canada N9A 6J5
Printed in the U.S.A.
© Copyright 2011 by International Business Machines (IBM) Corporation. This magazine could
contain technical inaccuracies or typographical errors. Also, illustrations contained herein may
show prototype equipment. Your system configuration may differ slightly. This magazine contains
small programs that are furnished by IBM as simple examples to provide an illustration. These
examples have not been thoroughly tested under all conditions. IBM, therefore, cannot guarantee
or imply reliability, serviceability, or function of these programs. All programs contained herein are
provided to you “AS IS.” IMPLIED WARRANTIES OF MERCHANTABILITY, NON-INFRINGEMENT
AND FITNESS FOR A PARTICULAR PURPOSE ARE EXPRESSLY DISCLAIMED. All rights reserved.
All customer examples cited represent the results achieved by some customers who used
IBM products. Actual environmental costs and performance characteristics will vary depend-
ing on individual customer configurations and conditions. Information concerning non-IBM
products was obtained from the products’ suppliers. Questions on their capabilities should be
addressed with the suppliers.
All statements regarding IBM’s future direction and intent are subject to change or withdrawal
without notice and represent goals and objectives only. The articles in this magazine
represent the views of the authors and are not necessarily those of IBM.
The following are trademarks (marked with an *) of the International Business Machines
Corporation in the United States and/or other countries. A complete list of IBM Trademarks is
available online (www.ibm.com/legal/copytrade.shtml).
.41�5FDI.FEJB�t�����4���UI�4U���4VJUF������.JOOFBQPMJT��./������t������ ������
To apply for or cancel your IBM Systems Magazine subscription, visit www.ibmsystemsmag.com.
. " 3 $ ) � " 1 3 * - � � � � � i b m s y s t e m s m a g . c o m / m a i n f r a m e10
Inspired by an eWeek article “Weird Job Interview
Questions,” (http://bit.ly/oddquestion) IBM Systems
Magazine asked its contributors:
What’s the strangest interview question you’ve been asked?
If I hired you and you
moved here, what would
your husband do for work?
Can you name five uses
for a paperclip?
Are you a fast driver? My answer was yes, naturally.
If you were a bird, what kind
would you be, and why?
How did they let you
get away with that?
Now availableWORLDWIDE
BIMONTHLY
DIGITALLY
and FREE!
Get the
CURRENT ISSUE NOW ibmsystemsmag.com/subscriptions
Communicating what’s next in IBM technology
PH
OTO
GR
AP
H B
Y T
RA
VIS
AN
DER
SO
N
OK, I’ll admit that during the cold winter months I’m a bit of a couch potato. While other
Minnesotans are out snow shoeing, skiing or sledding, I’m under a blanket eating popcorn and watching television. However, there are only so many shows and movies I can watch before it gets monotonous. That’s why one of my new obsessions is on-demand movies. I subscribe to a service that allows me one DVD per month, and unlimited movies streaming though the Internet.
I recently discovered a new feature of this service. Based on the rating I give movies, the rental service predicts other movies I’d like. Most of the time, when I’ve seen a recommended movie, even the predicted rating (out of five stars) is spot on!
When reading the articles for this issue, I began to think about the technology enabling these
recommendations. It’s a perfect example of IBM’s approach to smarter analytics; it goes beyond collecting and responding to business information, and takes it one step further to analyze the information and predict future business patterns. Get the overview on this approach in the Think Smarter column on page 18. You’ll also find articles about SPSS offerings that help discover patterns in data (page 32) and the cost effectiveness of Cognos* on System z* servers (page 36). Based on my experience with previous issues, I think this is one you’ll rate at five stars.
Natalie Boike, Managing Editor
Training for Health
Kathy Konkel, author of the
cover story, spends her spare
time running and training for
triathlons. Three years ago she
couldn’t swim the length of a
pool, and now
can swim close
to a mile in
open water.
She also
completed her first marathon in
Chicago last year.
Not so Koi
On warm summer evenings, Jim
Utsler, IBM Systems Magazine’s
senior writer, and his wife like to
relax next to their koi pond. The
fish, which aren’t actually koi but of
the less-expensive comet variety,
survive the Michigan winter thanks
to a heater and bubbler that keep
the water open.
Predicting a Good Read
CONTRIBUTORS
Contact Natalie at [email protected]
M A R C H / A P R I L 2 0 1 1 i b m s y s t e m s m a g . c o m / m a i n f r a m e12
Editor’s Desk
Inside this
month’s issue
M A R C H / A P R I L 2 0 1 1 i b m s y s t e m s m a g . c o m / m a i n f r a m e14
Dashboard News
to use
Charging AheadBefore electric vehicle sales can take
off, auto companies must address the
roadblocks revealed in a 2010 IBM
study: performance, recharging and
convenience. The IBM Institute for
Business Value surveyed 1,700 U.S.
drivers and interviewed more than
100 auto industry executives. Only
19 percent of drivers said they would
consider purchasing an electric-only
vehicle when shopping for a new car.
Source: IBM
Facebook use continues to exceed its own records. In November,
online intelligence company Hitwise reported Facebook.com
generates nearly one in four U.S. page views. Just eight months
prior, Facebook page views overtook Google.com visits.
Source: Hitwise Weblogs
Patent Not PendingIBM inventors received nearly 6,000 U.S. patents in 2010, topping the list of
the world’s most innovative companies for the 18th consecutive year.
Patents range from a new method for processing patient information to provide more
effective diagnoses, to computer communication via pulses of light. IBM invests
approximately $6 billion in research and development annually.
Source: IBM
Predicting Traffi c JamsIBM has partnered with the New Jersey
Turnpike Authority to provide smarter traffic
data to drivers. Instead of simple real-time
reporting of accidents, bottlenecks and
dangerous conditions, IBM software will predict
traffic patterns up to an hour in advance for
state roadways, including the New Jersey
Turnpike and Garden State Parkway.
Source: New York Times
Albuquerque, N.M. population 521,999
A lbuquerque uses an IBM Cognos
business-intelligence solution to help
its 7,000 employees collect data and
share information among more than
20 departments, from public safety to
libraries. The solution, which officials say
has cut administrative costs by about 2,000
percent, also helps bring information to
residents through an extranet.
Dubuque, Iowa population 57,250
IBM is building a platform that will give
Dubuque a real-time view of energy
consumed by the electric grid, water
system and general city services. The
city is also augmenting water meters
with a device that increases accuracy
and helps consumers identify waste,
and will offer financial incentives to fix
leaks. Policy makers and consumers will
benefit from a new services system that
IBM Research created to provide data
management and analytics.
In 1900, only 13 percent of humans lived in cities. By 2008, cities had attracted
half of the world’s population, and in 2050, 70 percent of us will be urbanites.
Our growing congregation in cities, their relatively manageable scale and their
adaptability to local needs make them perfect testing grounds for instrumented
systems, sustainable energy initiatives and data-enhanced services. Here are
some of the ambitious ways cities are leading us into the future.
PUBLIC SAFETY ENERGY AND UTILITIES HEALTHCARE GOVERNMENT SERVICES WATER TRANSPORTATION
50,000-100,000
100,000-250,000
250,000-500,000
500,000-1 million
7 million-8 million
10 million +
Smarter Cities L A R G E A N D S M A L L
CITY POPULATION
Chesapeake, Va. population 225,255
As part of a citywide capital-improvement
project, IBM will help Chesapeake enhance
services ranging from maintaining and
operating traffic signals and water systems
to managing police and fire departments.
A large city in terms of area, Chesapeake
has more miles of deep-water canals than
any other U.S. city, making it complex to
manage. Technology will help address this
complexity by collecting and analyzing
data to make transpor tat ion, ut i l i t y
management and public safety systems
adapt to dynamic conditions.
Sacramento, Calif. population 463,794
Wastewater agencies in California’s capital
can predict and solve equipment problems
before they affect citizens and businesses.
IBM software manages 98 pump stations,
3,000 miles of mainlines and 279,000
service connections with help from a
mapping system, control meters and live
video of the sewer pipes.
Data Display Visualizing
technology
M A R C H / A P R I L 2 0 1 1 i b m s y s t e m s m a g . c o m / m a i n f r a m e16
Sao Paulo, Brazil population 11,037,593
In the Paraguay-Paraná river system, which
provides drinking water for Sao Paulo, IBM
researchers and the Nature Conservancy
are running 3-D simulations that model how
land-use and water-policy scenarios will
affect biodiversity. The goal is to protect the
river basin, and therefore the water supply,
by making informed decisions.
Amsterdam, Netherlands population 762,057
In a pilot program, smart-metering systems
will enable 500 households to save on
energy costs and reduce CO2 emissions
by an estimated 14 percent. IBM, Cisco
and Dutch utility Nuon will develop home
energy-management applications and highly
secure data connections between household
appliances and the grid.
Bolzano, Italypopulation 102,880
Bolzano is in the midst of a one-year program
to test remote health monitoring for the
elderly. A group of 30 seniors is participating
in the Secure Living pilot to introduce
telemonitoring and teleassistance to the
homebound. A net of integrated sensors
will send data about the assisted person
to a control room, and the solution can
suggest tutored physical training, exercises
to stretch mental faculties or contacting
a health worker. If necessary, a medical
team is alerted.
Shenyang, Chinapopulation 7,760,000
Shenyang, an industrial hub with a
concentration of heavy manufacturing,
has an ambitious plan to address
environmental pressures. In a partnership
with IBM and Northeastern University, the
city has established the Shenyang Eco-City
Research Institute to develop technology
that helps cities cut carbon emissions,
conserve energy, manage water, track
food from farm to fork, smarten up
transportation and create environmental-
emergency response plans. For example,
the Institute will analyze a flood of real-
time data from the city’s water systems
to guide decisions about water quality and
energy used for water management.
Guangzhou, Chinapopulation 7,841,695
Guangzhou Metro Corporation, which
serves more than 2 million passengers
daily on the city’s rapid transit, aims
to make its system environmentally
sustainable and intelligent. Efficiency
depends on complete visibility across
assets to provide safe, uninterrupted
services. Guangzhou Metro has con-
tracted with IBM for a system to
manage its software, services and
physical assets including tracks,
carriages, advertising spots and sta-
tion shops.
Stockholm, Sweden population 825,057
After Swedes voted to reduce traf fic
through taxation, IBM built and installed
an advanced optical character-recognition
system to help the city charge drivers
who enter its congestion zone—an area
traditionally plagued by gridlock. Morning
traffic-queuing times are half what they used
to be, city traffic is down by 18 percent and
CO2 emissions are down 14-18 percent,
according to city officials.
Dublin, Irelandpopulation 505,739
Ireland’s Railway Procurement Agency
and IBM have integrated ticketing across
all public transport in greater Dublin. The
e-payment system, which can process up
to 2 million transactions a day, lets riders
use one prepaid card on all buses, trains,
trams and coaches.
Source: IBM
Taunton Deane, U.K. population 108,700
The Taunton Deane Borough Council is part
of a rare three-agency partnership to improve
service delivery and upgrade technology.
Southwest One—a joint venture of Taunton
Deane, Somerset County, and Avon and
Somerset Police—has enlisted IBM’s help
in ambitious projects that none of the three
entities could have achieved on its own.
i b m s y s t e m s m a g . c o m / m a i n f r a m e M A R C H / A P R I L 2 0 1 1 17
M A R C H / A P R I L 2 0 1 1 i b m s y s t e m s m a g . c o m / m a i n f r a m e18
Picturing Success
With business analytics, the more information you have, the clearer you see,” says Bernie Spang, director of strategy and marketing, IBM Database
Software and Systems. The explosive growth of information today—resulting from intelligent devices, sensors, cameras, the Internet and more—is an opportunity to more clearly see the business opportunities to pursue, wasteful spending to cut and risks to manage or avoid.
Capture insight from
real-time operational
data with smart
analytics systems
By Natalie Boike
To leverage information and realize these benefits, companies must employ a successful analytics solution that includes software for business intelligence (BI), data mining, text analysis and cubing capabilities, which are all tightly integrated and optimized with the data warehousing and hardware foundation, Spang says. To accomplish this, IBM offers workload-optimized systems designed to transform information into real business insight in a time frame that matters to your organization. “In all of the cases I’ve heard, clients say they receive one, two or all three of the following benefits: lowering costs,
reducing or eliminating risk, and growing revenue,” he says.
The Smart Analytics System, available on Power Systems* servers, eX5 systems and System z* mainframes, combines InfoSphere* Warehouse and Cognos* BI software with IBM servers and storage to create tuned and balanced business-analytics solutions that can be easily customized for any industry and expanded to add data and user capacity. “It’s got IBM expertise built in,” Spang says. “We’ve applied warehousing, business intelligence, server and storage expertise from three generations of solutions and 15 years to create an optimal balance of
ThinkSmarter
Future-proofing
your environment
“
i b m s y s t e m s m a g . c o m / m a i n f r a m e J U L Y / A U G U S T 2 0 1 0 19
capabilities. This saves our clients from investing time and money to put those pieces together themselves.”
Components for Smart Analytics
A successful business-analytics solution must have three core competencies, Spang says.
� Reliable, high-performance
data management. Analytics requires data-warehouse software that delivers access to structured and unstructured information in real time. The InfoSphere Warehouse, powered by DB2*, not only manages the data but also speeds systems deployment and optimizes data delivery using data partitioning, compression, workload and performance management.
� Integrated and governed
information throughout the data
lifecycle. IBM solutions ensure you have trusted information feeding your analytics. “If you’re not sure of the quality and reliability of the information and whether or not it’s being handled properly, your results are in jeopardy,” Spang says.
� High-performance analytics. Once you’ve gathered and stored the data, you can begin using the knowledge it contains. Leveraging database-oriented data mining and text analytics with IBM’s Cognos 8 BI capabilities, executives, line managers, business analysts and IT staff can leverage the right information in a manner that’s right for their needs.
Picture Increased Business
Combined, these capabilities help businesses find actionable insights in mountains of data, such as how to respond to customer preferences or predict outcomes. For example, tracking purchase history via a loyalty membership program, companies can analyze client demographics and behavior along with market trends.
A retail company could use this information to not only create a direct-marketing campaign, but also to do predictive analytics on its success. Spang says, “With predictive analysis, you can identify what kind of response
rate and revenue you can expect, and make a decision that’s best for the marketing dollars.”
Picture Efficient Processes
While the role of analytics in decision making is well known, more companies are tapping information to improve business-process management (BPM).
Insight, or business analytics, and action, or BPM, are two sides of the same coin, Spang says. “The multiplying effect comes when these two capabilities are in the hands of the business analysts—who use an integrated discipline to continuously drive growth, reduce costs and increase business agility across the organization,” he adds.
8QfU�i_e�XUQbT�dXU�Rejj/DXU�4UcdY^QdY_^�j�S_]]e^Ydi�Yc�Wb_gY^W�Q^T�
G5�G1>D�I?E�3?>>53D�1>4�3?<<12?B1D5
gYdX�S_\\UQWeUc��RecY^Ucc�`Qbd^Ubc��92=Ubc�Q^T�UTeSQd_bc
B5359F5�
=QY^VbQ]U�^Ugc��Y^V_b]QdY_^�Q^T�UTeSQdY_^�
AE1<96I�1>4�25>569D
Vb_]�]U]RUb�TYcS_e^dc��c`USYQ\�_VVUbc�Q^T�UfU^dc�
FYcYd�ec�Y^�1^QXUY]�Qd�C81B5�2__dX�! )�
6URbeQbi�"(�=QbSX�"
?b��W_�XUbU�d_�bUWYcdUb*�ggg�TUcdY^QdY_^j�_bW
M A R C H / A P R I L 2 0 1 1 i b m s y s t e m s m a g . c o m / m a i n f r a m e20
IBM’s BPM suite contains a comprehensive set of collaborative, role-based capabilities that enable clients to model, simulate, execute, rapidly change, monitor and optimize core business processes. Software and services from across IBM create a workload-optimized system, available in three editions: IBM WebSphere* Dynamic Process Edition, IBM WebSphere Lombardi Edition and IBM FileNet* Business Process Manager.
“Extending BPM to BI users, companies can make information and insight readily actionable and become a more responsive, flexible organization. In retail, accurately pinpointing the right client enables marketers to create an optimal offer that can be deployed as an online retail campaign in minutes rather than days,” Spang says.
Spang outlines some common customer-adoption patterns to that leve- rage BPM for improved business results:
` Integrate historical and real-time processes
`�Use data to drive process improvement
`�Link process execution with financial results for end-to-end optimization
`�Deliver BI at key decision points within operational processes to improve and automate decision making
`�Respond more quickly and effectively to changing market conditions
`�Provide complete business visibility—both insight and action
Picture Reduced Risk
Companies can also manage and prevent risk using fraud-detection offerings like InfoSphere Identity Insight. Spang explains the technology to analyze information about identities, people or businesses originated in Las Vegas, where casinos used it to screen employees before hiring them.
Identity Insight helps companies determine accurate identities and
uncovers relationships among identities. It can apply complex event processing to that knowledge to evaluate all of the transactions related to those individuals or businesses. Spang says IBM social-service clients, for example, use it to better understand their citizens and thus offer better services. Retailers use it to identify fraudulent claims.
Develop Insight in Real Time
Performance of these analytics systems is key in two areas. People often think of the warehouse performance and access to data. Indeed, systems must handle the ever-growing volume of information, the number of users and the types of analysis those users are demanding, Spang says.
Companies must efficiently deploy and maintain these solutions. Spang said one of his first experiences was with a healthcare insurance provider that complained of performance issues. “I studied up on feeds and speeds of the Innovative
Information Analytics
IBM has a long history with analytics. Here are a few of the milestones:
An IBM researcher invents
the relational database in
a project called System R,
which was designed to find
answers from data.
Think Smarter
1969
IBM creates the unstructured
information management
architecture, which is today
a standard for analyzing
unstructured data.
2004 IBM introduces InfoSphere
BigInsights, which leverages
software from the open-
source Apache project
Hadoop to gather and
analyze the vast amounts of
structured and unstructured
information available online.
2010 System S, an IBM Research
project for analyzing streams
of information, is brought
to market as InfoSphere*
Streams, analyzing data as it
flows through the system.
2008
Today’s technology could be
considered the third generation of
business intelligence, says Bernie
Spang, director of strategy and
marketing, IBM Database Software
and Systems. The first generation
was marked by creating data
warehouses to bring data together
and run reports for relatively few
business leaders. Next, people
wanted more information from their
data. Online analytics programming
enabled business analysts to
interact with the data and ask
questions of it, in addition to
creating the standard reports.
Now is the generation of
operational analytics, he explains.
Information can be empowering and
moves the focus from sense and
respond to predict and act.
“You want to enable everyone in your
organization, even your partners and
customers, to have greater insight
involved in every transaction, in every
business process, as it’s happening,”
Spang adds.
—N.B.
i b m s y s t e m s m a g . c o m / m a i n f r a m e M A R C H / A P R I L 2 0 1 1 21
Natalie Boike is managing
editor for IBM Systems Magazine,
Mainframe edition.
No End in Sight
Spang says as the increase in information continues, so does the appetite for analytics. “As technology has advanced, we’ve made it easier for more business users to analyze more information and apply more kinds of analytics than ever before,” he explains. “The more they get, the more they want: more data, more types of analysis. I see this as the relative beginning of an explosive growth of applying analytics to improve business results.”
software and all of our great offerings for high-performance analytics,” he says. “When I got there the CIO was talking about the performance of his teams and the ability to respond to business needs fast enough.” The business was acquiring new companies with data that needed to be folded in and analyzed more quickly. The more the users had analytics, the more questions they had.
With a solid base in high-performing IBM servers, storage, BI and data warehousing software, the Smart Analytics System meets these key performance requirements. “We’re working to integrate our software with our hardware and into tuned and balanced systems,” Spang says.
Unmatched Integration
and Experience
The scope and depth of the IBM portfolio and IBM’s holistic approach to data analytics—from collection and management, to integration and analysis—sets it apart from other offerings. IBM delivers smarter systems that are integrated and optimized from the processor to the application and does so based on a deep understanding of business needs and the workloads that support the organization.
“When we work with clients, we work to solve their most pressing problem, but we do it with an eye to where they ultimately want to go,” Spang says.
IBM BPM Suite: www.ibm.com/software/info/bpm/offerings.html
IBM Smart Analytics System: www.ibm.com/software/data/infosphere/smart-analytics-system
M A R C H / A P R I L 2 0 1 1 i b m s y s t e m s m a g . c o m / m a i n f r a m e22
Transitioning Workforce
One semi-retired guy is the only person I have who can pull us from a failure of one of our most important applications.”
This statement, from an application support leader at a North American bank, drives home the realities many mainframe-centric businesses face, whether business leaders realize it or not. While the media, hardware vendors and modernization service providers theorize about a shortage of mainframe workers, IT leaders responsible for meeting the world’s most stringent service level agreements (SLAs) face real situations everyday with only a limited number stop-gap fixes left, and no ground-level plan for implementing long-term fixes.
The Established Barriers
Mainframes, and the time-tested applications they run, provide
differentiating value to the world’s largest businesses. Even if it were possible to relocate these applications to another platform, it would take a brave (some would say crazy) IT leader to move away from the security and reliability of the mainframe to perform the massive number of transactions a single system accommodates on a daily basis—often well above the 20 million-per-day mark.
The reality is, it isn’t that easy. Most legacy applications can’t just be relocated and magically run in a new environment. They often must be rewritten, modernized or even scrapped altogether. This takes resources away from developing new revenue-generating capabilities, which
InsiderInsights from
the industry
Agnostic development
tools support
mainframe-centric
businesses
By Rose Rowe
“
i b m s y s t e m s m a g . c o m / m a i n f r a m e M A R C H / A P R I L 2 0 1 1 23
in today’s economic climate just isn’t good business.
The good news is mainframe computing is changing to accommodate evolving business practices. In the spirit of “if you can’t beat ‘em, join ‘em,” the mainframe, in particular the IBM zEnterprise* System, has expanded its architecture to accommodate customers who must provide services using many delivery platforms while maintaining the level of service the end user has grown to expect.
The bad news is perceptions remain a barrier to a full-steam-ahead strategy in IT organizations. Outdated perceptions about mainframe computing and its interoperability with other systems often deter senior managers from considering strategies to develop new services outward from the core business applications.
These barriers often leave application support leaders stranded somewhere between meeting daily SLA requirements and the timely delivery of new capabilities—with only a dwindling staff that has the knowledge base to do the work. Many options exist to work through these barriers, real and perceived, to enable businesses to expand their services through a technology strategy that will also enable application support leaders to meet those daily requirements.
A Platform-Agnostic
Perspective
This is already a reality for non-IT line-of-business staff. They just want new services on the devices they prefer. Whether it’s a mobile device, a website or a more traditional service, the business user wants quality and
dependability, regardless of how IT delivers it.
Fortunately, it’s becoming easier for IT to work on multiple platforms. Consider the following: any business that relies on a mainframe to deliver mission-critical services to its customers is likely storing 80 percent or more of its data on the mainframe. With Web-oriented architectures (WOAs) and service oriented architecture (SOAs), newer programming languages can easily integrate data on the mainframe with application logic on other platforms. With true platform-agnostic capabilities, the focus can turn to maximizing ease of use and accessibility for line-of-business users.
The same ease-of-use principles can be applied to application development. Programming tools can
DTS Software, Inc. announces the introduction of the MDL Control Center (MCC), a new product to facilitate the use of virtual tape systems from Bus-Tech® and EMC®.
MCC provides a wide range of components that allow installations to more effectively install, manage, and use the Mainframe Data Library.
1 Robust command and monitoring interface1 Advanced intelligent device selection1 Automated migration of tape libraries1 I/O load optimization
DTS Software – the leader in Storage Management, System Monitoring and Automation.
To learn more about improving your VTL efficiency, contact DTS Software at [email protected] or 770-922-2444.
FOR MAINFRAME VTL
EFFICIENCY USE...
MDL CONTROL CENTER
Visit us at SHARE
Booth #410
MDL is a product of Bus-Tech, Inc/EMC Corporation
M A R C H / A P R I L 2 0 1 1 i b m s y s t e m s m a g . c o m / m a i n f r a m e24
Insider
service and not on execution, and the time to market for the new capability is greatly reduced.
Getting Started
Implementing flexible environments that enable development on any platform is a pain-free start for organizations. In the recent past, organizations were faced with an either/or choice of development environments. The choice of environment either deterred nonmainframe developers or reduced the efficiency of its seasoned mainframe-development teams.
Today, environment isn’t such a concern. Many best-of-breed tools are available through open-source development environments, which are familiar to seasoned professionals and new developers alike. The flexibility of being able to choose a character-based green screen also ensures an organization’s most senior staff aren’t hindered.
For most developers, learning to develop in multiple languages is nothing new. Ask that same developer to use a new environment (especially a character-based environment), and not only is it no longer appealing to do the job, but the developer’s ability to do the job will also be hindered. With the no-cost option of open source and the flexibility to choose environments, addressing the development environment is an easy first step.
Enabling this cross-platform data manipulation is essential for mainframe-centric organizations. Yes, 80 percent or more of their data is likely
platforms, hardware and applications that deliver those services. Instead of having individual experts who understand the logic of specific applications or who work entirely on a single platform, IT can develop teams of experts who understand the logic used to deliver the service as a whole. Put this into practical operations and it will reduce time to market for new services.
For example, consider a business line manager who wants to include loan applications in the bank’s mobile banking services. In today’s operations, this request would be divided among three platform teams and multiple application development teams or individuals. Numerous project meetings and many testing points would be necessary to ensure these disparate teams are in step with each other as independent development decisions are implemented. While these checkpoints are necessary to ensure a quality end product, the business line manager has little insight into the true progress of the service being developed and may be dismayed by the length of time from request to implementation.
With platform-agnostic development teams, checkpoints are reduced to inner team collaboration. Development decisions can take each platform and application into account from within the team. This helps the team remain focused on the service itself as the endpoint. Ultimately, the business line manager has greater insight into the project, the development team is focused on
be a barrier to attracting top talent to work on applications that are important to the business. Traditional character-based green screens used for mainframe development are unappealing to the uninitiated for many reasons, starting with the drain on productivity as developers who are new to it stumble through its perceived limitations.
Modern development environ-ments that invoke best-of-breed tools, regardless of platform, are available to reduce (and possibly eliminate) these productivity concerns and learning curves. These environments allow a new generation of developers and progressive veterans to work on a business’s most important applications with minimal friction. They also enable the application support leader to task existing resources to shadow the partially retired resource without concern for the added overhead of learning a legacy develepment environment.
Streamlining Development
In the future, IT leaders can focus on the capabilities and services delivered by technology rather than on the
Many best-of-breed tools are
available through open-source development
environments, which are familiar to seasoned
professionals and new developers alike.
than invest limited resources into IT projects with no resulting business gain, organizations should consider evolving with their workforce and enabling their IT resources to take a business view of technology-enabled services that will drive revenue and growth. The results will mean more services are brought to market more quickly and IT will be better aligned with business with fewer constraints on future growth.
i b m s y s t e m s m a g . c o m / m a i n f r a m e M A R C H / A P R I L 2 0 1 1 25
stored on the mainframe. But even this data is stored in various databases and different formats. Organizations spend a great deal of their resources gathering and storing data. It’s essential for the business line manager to utilize the data toward the success of a project without being delayed by its need to be converted or its inability to be utilized along with other forms of data. Development teams focused on services require access to this data quickly, using standard procedures for gathering the data to safely test their new services.
Lastly, it’s reasonable for organizations to insist that their application-development tools integrate code-analysis functionality. This delivers practical guidance and insights to the application staff working with the code as they need
Rose Rowe is the vice
president of mainframe
strategy for Compuware
and is responsible for the development
and market strategy of Compuware’s
mainframe solutions.
it. Even relatively inexperienced developers will have key insight into the function, structure and deficiencies of the undocumented programs they must soon maintain and enhance.
Nothing will quickly replace the experience of our “semi-retired guy.” In lieu of funding a position to shadow this professional, other developers require insight into the application’s functionality before they can efficiently support it or build upon it.
Removing Platform Constraints
It’s fair to say this article is written from a mainframe perspective, but ultimately there shouldn’t be a battle between platforms. While resources are tight and funding for business growth is scarce, enabling a transitioning workforce makes great sense. Rather
M A R C H / A P R I L 2 0 1 1 i b m s y s t e m s m a g . c o m / m a i n f r a m e26
Mind the Gap
As service oriented architectures (SOAs) have evolved, so have the technologies and languages that support these Web services. For a business to fully leverage
its data, it must be mindful of the gap between the records themselves and the services that rely on them.
The splintered infrastructures used to store, manage and expose data make it difficult to effectively use information to its full potential. The integration of WebSphere* DataPower * and the zEnterprise* System extend the security and integration capabilities to a multiplatform architecture—ultimately improving XML/Web services and SOA integrations, security and throughput, while lowering costs and time to market for new services.
IBM recognizes the importance of sharing and integrating data across platforms, says Gari Singh, IBM product manager of SOA Appliances. “XML is that interoperable dialect,
but most of data is still sorted and stored in different formats. DataPower transforms and exposes that content as XML from the native format—such as CICS*—so it can be implemented on the Web or given to consumers or systems.”
A Brief History
WebSphere DataPower has been meeting SOA, security and integration needs since IBM acquired DataPower and its XML-integration technologies in 2005. The WebSphere DataPower Integration Appliance XI50 for zEnterpise (XI50z), announced Feb. 12, extends its capabilities.
Trends Industry signals
to keep you in
the know
DataPower support for
zEnterprise extends
its multiplatform
capabilities
By Natalie Boike
PH
OTO
GR
AP
HY
BY
(LE
FT
) R
EA
LS
MIL
EY/
FLIC
KR
(R
IGH
T)
BR
OW
NP
AU
/FLIC
KR
i b m s y s t e m s m a g . c o m / m a i n f r a m e M A R C H / A P R I L 2 0 1 1 27
task. DataPower enables customers to process those requirements quickly and easily. A typical proof of concept can be completed in two to five days and implementations are very quick. He adds, “From a management perspective, the new DataPower XI50z is a self-contained box that can be managed like a network device.”
This secure, scalable, easy-to-deploy ESB allows customers to:
h�Grow infrastructure without increasing complexity
h�Protect applications and data structures
h�Securely and quickly Web-enable a System z environment
h�Utilize previously unconnected applications, information and people
h�Manage policies for mission-critical applications
Businesses can use DataPower to engage in emerging spaces like social networking, cloud computing and Software as a Service (SaaS) to reach new markets and expand reuse of legacy resources and data.
The Significance of System z
For many customers, the System z server is the system of record for data and systems because of the built-in features like performance, scalability, security and more, Singh says. With the introduction of the multiplatform
zEnterprise System, many other needs and different types of applications in the distributed world can also benefit from integrating with the System z server-housed information and applications.
“The zEnterprise System takes it to the next generation,” he says. “DataPower already adds great value by integrating logically with software running on System z. The zEnteprise System enables the integration of special-purpose optimizers. Now I’m able to manage my DataPower and other appliances just like they’re part of System z and get all of the great benefits of being able to connect that to the distributed world.”
The DataPower XI50z is designed to meet SOA security and integration needs; it’s about having an ESB that’s secure and provides integration, with a low total cost of ownership and high performance and throughput. “What this offering really allows you to do is truly expand capabilities of System z solutions out into the distributed world. It enables you to do more with less—fewer physical resources and with less management—and do so more quickly.”
At its inception, DataPower used native compilation technologies to optimize processing of XML data into HTML, wireless markup language (WML) and other markup languages. The offering was a hardware-based appliance optimized for handling XML processing, schema processing and transformation. IBM paired it with WebSphere Application Server to essentially provide a security gateway, offload XML processing and have a highly configurable, higher-performance enterprise service bus (ESB).
Singh says these capabilities are still in high demand. “Customers obviously want to do more with less. The architecture and feature needs are still there and the fact that we do it in the client’s form factor gives us manageability, performance and consolidation.”
Implementing DataPower
At the highest level, the greatest business benefit of the DataPower Integration Appliance is the time to market and time to value, Singh explains. For example, a bank may have customer data stored on System z* servers that should be exposed to the Web services for online banking, or to distributed systems for internal applications such as customer relationship management (CRM). Singh says numerous project requirements are necessary to accomplish either
DataPower is one of several high-performance specialty processors for
specific workloads available for the zEnterprise System. For additional information,
visit www.ibm.com/systems/z/hardware/zenterprise/index.html.
Natalie Boike is managing
editor for IBM Systems
Magazine, Mainframe edition.
Martin Kennedy, Citigroup Inc.’s
managing director of enterprise-
systems infrastructure, says the
mainframe provides the highest
levels of security.
i b m s y s t e m s m a g . c o m / m a i n f r a m e M A R C H / A P R I L 2 0 1 1 29
Whether for internal or external reasons, most organizations require data be backed up and available for disaster-
recovery or business-continuity purposes. In particular, financial institutions are often held to a higher standard because of the sensitive nature of the data they handle.
Although most customers “don’t see the detailed processing that happens in the background, the fact that we can tell them that their data is
being managed with the highest levels of integrity, availability and security is a big plus,” says Martin Kennedy, Citigroup Inc.’s managing director of enterprise-systems infrastructure.
That’s why Citi created a backup environment many would consider best-in-class, using four IBM System z10* Business Class (BC) machines as dedicated mechanisms for backing up the company’s myriad servers and systems. Combined with INNOVATION Data Processing’s FDR/UPSTREAM and UPSTREAM/SOS, their design not only ensures proper backups, but also optimizes CPU usage and network bandwidth.
“The traditional backup model limits backups to an off-hour window because of the high CPU and bandwidth requirements,” says Kennedy. “With our model—with the data moving through a dedicated SAN [storage area network] fabric and not over TCP/IP—you can pretty much schedule backups whenever you want without impacting production applications and workflow.”
A Sound Investment
Citigroup tames its
backup environment with
dedicated mainframes
PH
oto
gra
phy
by
Matt
Gre
ensla
de
UP CLOSE
M A R C H / A P R I L 2 0 1 1 i b m s y s t e m s m a g . c o m / m a i n f r a m e30
A Cost-Effective Mechanism
Citi is a global provider of consumer, corporate and investment-banking products and services. Its many brands include Citibank, CitiMortgage, Citi Cards and Women & Co. Citi operates in more than 160 countries and handles around 200 million customer accounts.
With such a large presence, you’d expect it to have a large IT infrastructure, and it does. It runs servers of all types, including IBM mainframes and Power Systems*, Linux* (stand-alone on System z*), Windows* and UNIX* OS-based machines. These boxes handle core functions: serving customer accounts, running business-intelligence applications or supporting internal back-office accounting functions. They all produce a huge amount of data, most of which is backed up, stored and ready for recovery for regulatory or internal purposes.
This data is vital to Citi’s everyday operations. That’s why the company put so much effort into creating a seamless data-retention model: from point-of-origin to vaulted storage. But things weren’t always so efficient, as Kennedy explains.
“Back in 1996, we were dealing with our first large data warehouse with around 20 terabytes of data, which was large for that time,” he says. “We would have to break backups into 20 separate components, 20 separate servers and 20 separate tape libraries. We only had one window in which to do that, typically on Sunday afternoons: the only time we could take a machine down for system backup. We struggled with that quite a bit.”
That’s probably an under-statement given the scale of the organization and the growth in the amount of data it was collecting. Network bandwidth, for example, became an issue, forcing Citi’s IT staff to tune the network continuously to accommodate increased traffic. This affected everyday operations as data
movement competed with typical application transactions.
“We revisited our backup architecture,” Kennedy recalls. “As a result, we considered using INNOVATION tools, including FDR/UPSTREAM and UPSTREAM/SOS, both of which would allow us to move data in a more robust environment. No sooner had we converted to EMC and INNOVATION than our issues related to backups almost literally disappeared.”
These improvements resulted in part because a SAN fabric and transfer devices had been introduced into the storage environment. Using this SAN, Citi IT staff could bypass the production network and run backups whenever they were scheduled, without worrying about a network bottleneck and reduced application response times. This also held true with its Linux on System z instances, with UPSTREAM/SOS passing these mainframe-
Headquarters: New York City
Business: Global provider of financial
products and services for consumers,
corporations, governments and institutions
Challenge: Improving data-backup
processing
Solution: Using four IBM System z10
Business Class machines as dedicated
backup servers and INNOVATION Data
Processing’s FDR/UPSTREAM and
FDRSOS to automate backups
Hardware: Four IBM System z10
Business Class servers
Software: INNOVATION Data
Processing’s FDR/UPSTREAM
and FDRSOS
based backups over System z FICON* and Fibre Channel Protocol (FCP) channels rather than over the network.
Ordinarily FDR/UPSTREAM pushes distributed backups to mainframe-based tape or DASD, but
CUSTOMER: Citigroup Inc.
they allowed Citi to use LPARs on its mainframe systems to channel backups. This meant the organization’s previously distributed backup model for its largest and most complex servers could be consolidated to fewer systems and therefore fewer backup media.
“This is a very cost-effective backup mechanism,” Kennedy says. “We were able to leverage the mainframe tape infrastructure, which allowed us to make the most of our cartridges by filling them with data that had like retention requirements.” That can be difficult to do in a distributed backup environment, where overall cartridge utilization of unexpired data tended to be lower. Managing cartridge utilization is key to managing the overall cost of a backup solution.
An Important Characteristic
Other organizations may have been satisfied with these improvements, but not Citi. When IBM announced the System z10 BC machines, Citi decided to further bolster its backup model by dedicating these boxes—one located at each of the corporation’s primary North American data centers—purely for backup purposes. This would allow the organization to move all of its backup loads for its largest servers off its production systems.
Citi installed one z10 BC at each of its main North American data centers, with the boxes supporting the local backup requirements for large servers at these data centers. When a group wants to add a server to the backup schedule, it fills out a change-management template that identifies what it wants backed up, how often they want it backed up and what the retention period is for the backup. At that point, someone in the distributed backup team creates an FDR/UPSTREAM job and schedules it according to user requirements.
Initial backups are of the full system. Following that, daily incremental backups take place, with FDR/UPSTREAM looking to copy only the files that have changed since the previous backup. Once a week, merged backups combine the incrementals to create a new, full-system weekly backup.
“Those weekly backups can take place outside the typical window for server backups, and we don’t have to be connected to the server to do that,” Kennedy says. “This is in contrast to many backup models, where you have to be connected to the server. Now we can drive those business class machines 24 hours a day with productive work without impacting the servers that are being managed by those machines. It’s all very efficient.”
Citi can now also aggregate backups based on retention. In the past, different backups with different retention rates were stacked on top of each other. Sometimes tapes were kept in storage because, even though one server’s retention-based backup may have expired, others hadn’t.
“You might end up with a situation where 60 or 70 percent of the data on a tape might be expired, but you couldn’t reuse that tape because the other 30 or 40 percent had not yet expired,” Kennedy says. “You ended up with tape capacity you couldn’t reclaim. Since FDR/UPSTREAM uses more traditional mainframe tape-management mechanisms, we can now stack data so when data on the tape has expired, the tape is immediately available for reuse.”
This data is encrypted as it’s written to tape, some 650 TB a month,
according to Citi’s Jim Jurasin, vice president of enterprise systems infrastructure. The encryption process occurs on the organization’s IBM tape drives instead of on the System z10 BC machines. This reduces CPU overhead, which allows the mainframes to run their jobs without taking performance hits. Citi has also
implemented disk encryption on the backup servers “to ensure the environment is completely secured from the time the data goes to disk and then to tape,” Kennedy adds.
“That’s an important characteristic for us,” he continues. “We have a lot of very sensitive data around here, so it’s critical that we make it as secure as possible, and IBM has some strong algorithms built into the machines, as well as crypto engines and certificate management.”
Exceeding Expectations
Citi’s customers probably don’t think about this type of back-end processing, but if they did, they’d be impressed at the measures the corporation has taken to make sure their data is backed up, safe and secure. Additionally, this streamlined backup model has also benefitted the corporation itself. Backups that run on the BC machines no longer bog down the corporation’s traditional networks, servers don’t have to be taken offline when they’re being backed up, backup media is being used more efficiently and automated processes have reduced backup administration. In the end, it’s all about Citi’s customers and their service expectations—which are now more than being met.
i b m s y s t e m s m a g . c o m / m a i n f r a m e M A R C H / A P R I L 2 0 1 1 31
Jim Utsler, IBM Systems
Magazine senior writer,
has been covering
technology for nearly 20 years.
“IBM has some strong algorithms built into
the machines, as well as
crypto engines and
certifi cate management.”
—
Managing director of enterprise-systems infrastructure, Citigroup Inc.
¨ Predictive analytics enable higher-
quality decisions, measurably better
outcomes and a higher ROI.
¨ Deploying on System z servers
ensures analytic processes are secure,
transparent and compliant.
¨ IBM provides an end-to-end
business analytics solution that can
scale to meet business needs.
TAKEAWAY:
IBM offers strong capabilities in information management, reporting and analysis. A merger with SPSS in 2009 further enables customers to drive competitive action from both structured and unstructured data. SPSS was an early driver of
predictive analytics and influenced its emergence on the market; now it’s an established leader in the field. This IBM company’s predictive-analytics offerings provide organizations a distinct advantage as analytics becomes a mainstay in today’s gridlocked marketplace.
The IBM SPSS predictive-analytics software portfolio combines various capabilities that integrate multiple data sources for statistical, mathematical and other algorithmic analyses and predictive modeling—along with an infrastructure that helps organizations effectively deploy predictions. The results are higher-quality
decisions, measurably better outcomes and a higher ROI.Business analytics combines the forward-looking
capacities of predictive analytics with the data-exploration and reporting capabilities of business-intelligence
applications. Because it gives organizations the power to use their rich stores of data in many different ways, business analytics is at the heart of providing business insight; it’s the engine that drives better outcomes.
The Real ROIOrganizations that invest in predictive analytics improve their capability to gain detailed insight into present conditions and to evaluate likely future events and outcomes. They quickly identify ways to improve business performance by cutting costs, minimizing risk and developing successful strategies for increasing revenue. They often outperform their peers. Not surprisingly, the demand for predictive analytics continues to grow. In a 2009 IBM study, 83 percent of CIOs said analytics is a priority.
Companies that deploy predictive solutions clearly demonstrate the power of predictive analytics. Ninety-four percent of SPSS customers achieved a positive ROI with an average payback period of 10.7 months, according to a Nucleus Research study. Returns were achieved through reduced costs, increased productivity, increased employee and customer satisfaction, and greater visibility. Flexibility, performance and price were all key factors in SPSS software purchase decisions.
i b m s y s t e m s m a g . c o m / m a i n f r a m e M A R C H / A P R I L 2 0 1 1 33
Successful companies fuel business with predictive analytics
By Kathy Konkel
Illustration by Oliver Burston
Applying AnalyticsMost commercial organizations share similar goals: attract the best, most profitable customers; grow the business through up- and cross-sell; retain high-quality customers; minimize risk to the organization; and detect and prevent fraudulent activity. Supply-chain considerations, such as effective assortment planning in retail stores and efficient production lines in manufacturing, are also key considerations that can significantly affect the customer experience and ultimately customer value. In short, the goal is to grow revenues and minimize costs, thereby producing profitable revenue growth. Typical applications include:
h�Delivering well-targeted campaigns
h�Classifying high-quality customers and, conversely, identifying those who are costly
h�Focusing investigators and minimizing the effect of fraudulent activity
h�Leveraging effective and reliable inventory management
h�Predicting maintenance needs
Public agencies manage a unique set of challenges and often face additional scrutiny because of the personal nature of public security, healthcare and education. Because they’re also often funded or subsidized by government-run agencies, these organizations face additional pressures and bureaucracy around operational budgets. Typical public-sector applications include:
h�Government agencies that manage functions as diverse as tax audit selections, military force recruitment, and proactive policing and public safety.
h�Healthcare organizations that seek to proactively manage their resources and fine-tune their practices to provide better patient care.
h�Colleges and universities that manage the entire student lifecycle: recruiting the right mix of students, offering a selection of programs and assistance to keep students enrolled, and managing alumni development programs.
Data-Driven DecisionsAt the highest level, predictive analytics enables better business decisions. Analysis might reveal new insights that help senior management drive far-reaching strategic decisions and deliver step changes in business value. However, these insights are more often applied at the individual-case level, enhancing key business decisions that are made frequently and repeatedly, where improvement leads to a higher proportion of good outcomes and clearly measurable, incremental ROI. A helpful way to think of these decisions is: If we could make better decisions about “X,” we could deliver greater value by doing “Y.” For example:
h�If we could reliably predict which of our high-net-worth customers were likely to defect to a competitor, we could ensure their continued business by offering incentives.
h�If we knew how likely each of our customers would be to respond to a particular cross-sell offer, we could reduce the size and cost of campaigns (and increase response rates and revenues) by not targeting people unlikely to respond.
h�If we could accurately assess the risk of each insurance claim as we receive it, we could reduce costs and increase customer satisfaction and loyalty by fast-tracking safe claims and increase our fraud-detection rate by
ensuring our investigative resources focus on genuinely high-risk cases.
Integrating the results of analytics with business processes and operational systems, or deployment, can be relatively simple: one point of integration in one process and system. For example, at a single point in processing tax returns, a predictive model scores every return on the likelihood of noncompliance, and adds those with high scores to an investigations-team audit list. In other solutions, the approach might be more complex.
Why System z?While organizations strive to deliver analytics that will make decision making smarter, better and faster, they also must ensure that analytic processes are secure, transparent and compliant with external and internal governance regulations. While information infrastructures must be high-performing, management strives to make them less costly.
IBM recently released versions of the SPSS predictive analytics solutions designed specifically for IBM System z* environments. By running analytics on System z servers, you can deliver sharp and timely insights that everyone, from senior executives to frontline managers, needs. You can also meet requirements for expanded data-accessibility and security, process transparency, information-systems efficiency and cost savings.
Predictive Analytics
n. Predictive analytics draws
reliable conclusions about current
conditions and future events and
thus helps connect data to effective
action. It enables organizations
to make predictions and then
proactively take action based
on that insight to drive better
business outcomes and achieve a
measurable competitive advantage.
To support the entire analytical process, the SPSS solution for System z* consists of the following components:
IBM SPSS Modeler for Linux* on System z is a powerful, versatile data and
text-analytics workbench that helps analysts build accurate predictive models
quickly and intuitively, without programming. Using leading data-mining
techniques, the workbench helps analysts model future customer behavior.
It’s the most efficient way for organizations to use their data to model future
customer behavior. This technique discovers patterns and trends in structured
or unstructured data more easily, using a visual interface. In addition to
predicting outcomes, models can explain the factors that influence them so
users can take advantage of opportunities and mitigate risks. Text-analysis
capabilities are integrated with SPSS Modeler too, so the critical insights
locked in unstructured data can be used to increase the accuracy of
predictions about customer behavior and strengthen the modeling capability.
IBM SPSS Statistics for Linux on System z is a leading statistical software
that commercial, government and academic organizations use to solve
business and research problems. IBM SPSS Statistics is one of the most
accessible statistics tools in the market, enabling organizations to apply
mathematical discipline to their decision-making.
IBM SPSS Collaboration and Deployment Services for Linux on
System z provides a platform for managing and deploying analytical assets
and allows for secure, browser-based access to results and integration to
other components. It provides an infrastructure for managing the storage and
automated refresh of predictive models along with other analytics processes.
In addition to predictive-analytics solutions, IBM Cognos* 8 Business
Intelligence is also available for Linux on System z. You gain the real-time
monitoring, reporting, analysis and dashboards it provides—plus all of the
performance, efficiency and security of a System z environment. More
information is online (www.ibm.com/software/data/cognos/cognos-8-
business-intelligence/system-z/).
The System z server supports both predictive analytics and business-intelligence solutions. It simplifies and provides faster access to the data it holds without impacting transactional-system performance. Its virtualization environments enable organizations to allocate resources either as shared resources or fixed-resource assignments, which ensures the system can meet the needs of all users, regardless of the number of processors required. Consistent response time is also required as predictive models are used to make recommendations within critical business processes to improve decisions at the point of impact. Read “System Components” (right) to learn how SPSS and System z solutions integrate.
Additionally, System z servers provide the highest level of data security, with features built into the OS and the hardware. This is critical when the data being used for analytics is sensitive and critical to the business. It also provides five nines of availability, with automatic failure or potential-failure detection with built-in hardware and OS recovery. With minimal incremental costs, your predictive-analytics solution is available when and where you need it, with less strain on your network and your IT staff.
With SPSS under its umbrella, IBM provides a true end-to-end business analytics solution that can scale to meet your evolving needs. Benefits include increased ROI, higher user satisfaction, industry-leading scalability, reliability, availability and security in a flexible infrastructure, and ensured security and enforcement of regulations and standards.
i b m s y s t e m s m a g . c o m / m a i n f r a m e M A R C H / A P R I L 2 0 1 1 35
Kathy Konkel is a member
of the IBM SPSS product
marketing team and has
worked with the SPSS products for more
than 10 years. She’s currently responsible
for supporting the enterprise sales channels
for SPSS, which includes specialized
software sellers throughout IBM.
¨ IBM studied costs based on
acquisition, maintenance, facilities and
administration expenses.
¨ Regardless of deployment size,
the System z platform provides a more
cost-effective alternative to a distributed
infrastructure.
¨ The IBM mainframe deployment pays
for itself in five years, based on system-
administration cost savings alone.
TAKEAWAY:
i b m s y s t e m s m a g . c o m / m a i n f r a m e M A R C H / A P R I L 2 0 1 1 37
hat would you think of reducing your cost of delivering business
intelligence (BI) by upwards of 50 percent over the next five years? What would you say to allocating a smaller portion of your IT budget to facilities and administrative costs while providing BI to a broader audience?
These questions come at a pivotal point for many organizations. The consistent gap between the promise of BI and the harsh realities of today’s BI implementations is widely recognized. Many deployment models limit the capability to provide BI access to all potential users with the performance they’ve come to expect from an enterprise application.
Perceived Barriers
Creating a successful BI strategy and providing the supporting infrastructure isn’t a trivial undertaking. To date, many IT organizations have implemented isolated BI solutions in individual departments in lieu of an enterprise model. Little focus has been put on developing a BI standard for the whole business, resulting in tool overlap, shelf ware, duplicate license agreements, underlicensing exposure and technology gaps.
In addition, the floor space, cooling, system administration, acquisition and maintenance contract costs needed to support these multiple BI solutions can become cost prohibitive over time. After making major investments in BI, organizations have also found that users eventually dropped it because of inadequate performance, reliability, data quality or security assurance.
With this in mind, IBM set out to understand the actual costs of delivering BI to businesses in a manner that would both meet and exceed expectations.
Collecting Data
IBM wanted to understand how costs varied over five years for various BI deployment sizes (100, 1,000, 10,000, 20,000 and 50,000 named Cognos* users), while striving to answer four key questions that most organizations ask themselves when defining a BI strategy:
When adhering to industry best practices, what are the
total cost of ownership (TCO) and total cost of acquisition (TCA) of the BI infrastructure and what implementation options are available?
What happens to TCO when hardware is replaced every 36
to 60 months (the common industry practice)?
What impact will growing BI requirements have on
infrastructure expenses?
How much will it cost to ensure ongoing availability of a mission-
critical BI solution?
The study’s goal was to understand what BI infrastructure options are available to companies for meeting their business’ strategic performance
IBM demonstrates Cognos BI on System z is less expensive than x86 deploymentsBy Rebecca Wormleighton
ILLU
STR
ATIO
N B
Y M
ICH
AEL
AU
STIN
M A R C H / A P R I L 2 0 1 1 i b m s y s t e m s m a g . c o m / m a i n f r a m e38
expectations and lowering costs. The next step was to compare the TCO and TCA of IBM Cognos BI on IBM System z* to x86 distributed servers.
The Assessed Costs
All costs used in the model were based on realistic defaults developed from North American and European business enterprises. The study compared the costs associated with deploying Cognos BI on nonvirtualized, distributed, rack-mounted x86 servers with quad-core 2.26 GHz Intel* E5520 Xeon* technology to System z10* Enterprise Class servers running Linux* OS.
The Right-fitting Applications into Consolidated Environments (RACEv) tool (where v stands for virtualization), designed by the IBM Systems and Technology Group, helps evaluate the following categories when exploring the TCO of the various platform options available:
h�Acquisition This includes costs associated with purchasing everything necessary to deliver BI to the business, including hardware, network-connectivity equipment and software (excluding the Cognos BI software, because it’s the same for all cases with named user pricing), along with the additional hardware, software and connectivity required to ensure high availability of the BI solution.
h�Maintenance agreements Like acquisition costs, maintenance agreements are an unavoidable cost of any IT initiative.
organizations that are considering the future costs of technology refresh, growth and high availability.
One of the most striking findings, uncovered with the RACEv tool is the System z solution pays for itself over five years based on the savings from system-administrative costs alone. While initial acquisition costs may be marginally higher, after five years, the TCO is consistently lower.
With the introduction of IBM business analytics on System z (see “Driving Outcomes,” page 32), IBM can offer a solution that takes BI a step further—on a single platform that scales to meet the need for faster and better access to complete and accurate business with less resources and expense. A flexible solution, it can help companies address the business challenges of today and the evolving business needs for actionable insights that help to optimize business performance.
Rebecca Wormleighton is
an IBM product marketing
manager for Cognos
software, focusing on
synergies between Cognos business-
intelligence software and IBM products.
Read the complete study, “The New Alternative for Leveraging the Power of Business Intelligence,” online: http://bit.ly/cognosonz
Although it may be possible to negotiate varying levels of discounts, the vendor generally predefines the costs. Maintenance contracts, like the products themselves, come with a defined list of features, and companies must weigh the value of these features to their business and consider the costs of accessing those features elsewhere.
h�Facilities These costs include floor space and power costs for BI infrastructure and represent, on average, approximately 10 percent or more of a company’s overall operating costs. This is an area that can easily be forgotten when exploring TCO. If not monitored, facilities costs can quickly spiral out of control. This especially holds true when a BI infrastructure is built out for various departments without taking into account the available capacity that might already be available.
h�Administration This category covers the costs to deploy and operate a BI infrastructure design based on stated best practices and includes the resources for installing, supporting and maintaining that infrastructure—while planning for and responding to service outages and other problems.
System z Costs LessThe study concluded cost barriers associated with a System z implementation versus x86 are perceived, not real. The acquisition costs (hardware, network and software) aren’t the barrier to entry as people believe. The study found that the System z platform offers a powerful, more cost-effective alternative to a distributed infrastructure, regardless of deployment size.
In addition, inherent System z capabilities provide stronger performance, better security and greater scalability. Cognos BI on System z enables organizations to help ensure the high availability of their BI solution at a price they can afford; TCO is dramatically lower for
Learn from product developers and industry experts. Expand your IT education by attending one of the many technical conferences on our schedule for 2011. IBM conferences offer in-depth information on recent technical product announcements, tips and techniques you can apply immediately, hands-on labs, discounted certifi cation testing at select events, and provide a great opportunity to network with others.
A smarter planet starts
with you
Build your skills with IBM Training
2011 IBM Technical Conferences (U.S.)
DATE CONFERENCE LOCATION
REGISTER TODAY!ibm.com/training/conf/usAttention Exhibiting Sponsors: Exhibitor and sponsorship opportunities are available.For more information, visit our web site, click on the conference of your choice and then click on sponsor/exhibitor details.
April 10 - 15, 2011 Impact 2011 Las Vegas, Nevada April 27 - 29, 2011 IBM Power Systems Technical Symposium Orlando, Florida April 27 - 29, 2011 IBM System z Technical Symposium Orlando, Florida May 16 - 19, 2011 2011 IBM Exceptional Orlando, Florida Web Experience - Americas June 5 - 9, 2011 Innovate 2011 Orlando, Florida July 18 - 22, 2011 IBM System x Technical University Orlando, Florida July 18 - 22, 2011 IBM System Storage Technical University Orlando, Florida October 3 - 7, 2011 IBM System z Technical University Miami, Florida October 10 - 14, 2011 2011 IBM Power Systems Miami, Florida Technical University October 23 - 27, 2011 Information On Demand 2011 Las Vegas, Nevada
Early Bird registration discounts are available for many conferences
Tips and Techniques
Tactical
solutions
M A R C H / A P R I L 2 0 1 1 i b m s y s t e m s m a g . c o m / m a i n f r a m e40
Optimal Service Delivery
Managing today’s complex server environments, networks and IT infrastructures is challenging. Most organizations have tools and disciplines in
place to provide basic availability management, including hardware and software monitoring capabilities. However,
IT staff may not have the skills or time to support and implement notification handling, performance and capacity management, security management, configuration management and automation. With the pressure to do more with less, how can an organization balance the need to drive down costs and improve delivery of system administration, information access, remote access, and problem and change management?
This article surveys the challenges and solutions for effective management of your server infrastructures and network devices, as well as important disciplines such as availability management and the tools you need to support a complex IT environment.
Understanding Challenges
An infrastructure typically consists of a broadly heterogeneous collection of elements such as servers, virtual machines, logical partitions, routers, switches, wireless appliances, disks, processors, operating systems, middleware and database instances, and
batch jobs. All of these elements can be mapped, more or less, into a simple framework as shown in Figure 1 (page 41).
The management of an organization’s heterogeneous and complex IT infrastructure is often a significant portion of your total cost of ownership (TCO). Getting costs under control while also meeting your service-level commitments requires a focus on the following eight areas of infrastructure-management challenges:
h�Detect and handle incidents and problems.
h�Handle changes with minimal impact on availability.
h�Prevent security problems. h�Effectively implement emerging
or challenging technologies. h�Maintain server software and
firmware. h�Employ useful reporting
indicators. h�Supply the right tools. h�Rapidly deploy infrastructure
and tools with ongoing management.
Overcome eight
key challenges and
reduce costs
By Joseph Gulla
i b m s y s t e m s m a g . c o m / m a i n f r a m e M A R C H / A P R I L 2 0 1 1 41
Plan Effective Management
Organizations should focus on solutions that address these challenges as they evaluate their current infrastructure-management capabilities. A focus on value should shape your goals, with cost and complexity being major factors. You can then determine the best approach to achieving your goals in the context of specific tactics, as shown in Figure 2 (page 42). Consider the following recommendations to address the eight key challenges:
1Detect and handle incidents and problems. Incidents and problems
are a way of life for systems that involve hardware, software and people. But detecting and handling problems effectively isn’t easy. It requires discipline, imagination and experience. In addressing this challenge, the goal should be to provide robust incident and problem handling using preemptive actions and both proactive and passive monitoring. For example, a service-level agreement (SLA) could resolve 90 percent of Severity 1 problems in less than four hours. Assign support personnel based on the severity of the incident to improve response time, including a specialized service manager or multidisciplinary team if required.
2Handle changes with minimal impact on availability. Change is
frequently required in today’s highly complex systems. Dealing with change while striving for minimum downtime requires many skills. Meet weekly with team members to plan change management and estimate the time required as part of the preparation. Test changes prior to their implementation and prepare back-out activities in the event of a failed change.
3Prevent security problems.Security challenges are ongoing
and constantly changing. Security-management solutions should
incorporate plans to address threats to your systems and information in the form of both Web-based attacks and hackers. Balance security-risk management with maintaining business agility. Implement key security variables such as password length and update frequency early in the process and perform security remediation as required for servers and other devices to refresh and update your security posture. Look for security exposures during the ongoing support period. For a closed-end process, specify the frequency of these analyses with monthly reporting on key security attributes and activities related to servers
and other devices. In doing so, you’ll ultimately see reduced financial risk associated with security breaches.
Applications, services
Middleware
Operating system
Computer and network hardware
PL
AT
FO
RM
Software and hardware service layers in distributed systems
Figure 1
(Coulouris, Dollimore, & Kindberg, 2005)
Tips
M A R C H / A P R I L 2 0 1 1 i b m s y s t e m s m a g . c o m / m a i n f r a m e42
4Effectively implement emerging or challenging technologies.
New technologies hold great promise for businesses because they can save money and increase flexibility. Even so, the benefits of new technologies can be elusive when unfamiliar users are learning to use them. Focus on technologies such as virtualization and high-availability configurations. Virtualization provides flexibility to achieve rapid change and high utilization of your existing IT resources. High-availability technology supports business resiliency and your requirements for continuous application availability. Additionally, focus on skills support for virtual machines and logical partitions, using change windows to make dynamic changes to production servers—like adding processor or memory resources to an image. Develop the skills of your staff in many high-availability software solutions.
5Maintain server software and firmware. The optimal
data center would include a balanced collection of proactive and reactive activities supported by effective software products and tools. Startup tasks may involve refreshing operating-system processes, establishing startup sequences and changing priorities. Other server-support activities could include patching, configuration and log-file maintenance, using remote tools to perform operational procedures, server security activities for identification and access, and support for virtualization software. Applying emergency security fixes or even routine software changes can be a daunting task. Avoiding negative impact on application availability and performance requires planning and careful execution, including provisions for backing out failed updates.
6Employ useful reporting indicators. Information on the status
of systems and applications is important in infrastructure management, especially in problem situations. This information is also needed for capacity planning.
Activity or Discipline Tool Example
Availability management IBM Tivoli* NetView* for z/OS*
Hardware monitoring z/OS Hardware Management Console (HMC) and IBM Tivoli Netcool/OMNIbus
for inclusion in an end-to-end view
Software monitoring of resources IBM Tivoli OMEGAMON* XE for z/OS
Performance and capacity management OMEGAMON Product Suite for IMS*, DB2*, networks and storage
Security management IBM Tivoli Security Manager for z/OS
Configuration management IBM z/OS HCD Hardware Configuration Definition (HCD) and Hardware
Configuration Management (HCM)
System administration z/OS Utilities and SMP/E
Information access IBM Tivoli Business Service Manager for z/OS
Automation IBM Tivoli NetView for z/OS and Tivoli AF/OPERATOR on z/OS
Problem and change management IBM Tivoli Change and Configuration Management Database and IBM Tivoli
Application Dependency Discovery Manager
Tools in ContextTable 1
CO
NT
EX
TC
ON
TE
XT
Business
Alignment
Cost and
ComplexityGoal
What are
we trying to
achieve?
Tactics
What can we do
to achieve it?
VA
LU
E
Planning framework for an effective IT management solution.
Figure 2
Business
Alignment
i b m s y s t e m s m a g . c o m / m a i n f r a m e M A R C H / A P R I L 2 0 1 1 43
However, obtaining and analyzing operational and trend information is time-consuming and complex. Indicators should have dynamic reporting capabilities as well as monthly reports that provide a good way to understand the activities and performance of the service. An operational or business dashboard is also useful. Posting reports on a portal for easy access and using tools that focus on server resource management and generate performance and capacity reports will help analyze trends.
It’s also important to generate and post reports on security management. A portal can also be used to enter problem records and change notifications–while providing links to other needed tools, portlets and services that support ongoing activities. Dynamic reporting supports functional disciplines such as performance and capacity management.
7Supply the right tools. Having the right toolkit for your environment
requires balancing human aspects, types of tools, number of tools, level of integration and secure use. Appropriate tools and processes deliver optimal service and support your SLAs. Organize your tools in categories, including those that support key activities as shown in Table 1, page 42.
8Rapidly deploy infrastructure and tools with ongoing
management. The fast pace of change
The fast pace
of change in today’s
business environment
requires rapid sys-tems deployment
and tools that provide
ongoing support.
Tips
M A R C H / A P R I L 2 0 1 1 i b m s y s t e m s m a g . c o m / m a i n f r a m e44
infrastructure and network devices. By focusing on the disciplines and tools necessary to carry support and delivery services, you can improve your own management capacity and evaluate the capabilities of potential outsourcing and managed-service providers.
in today’s business environment requires rapid systems deployment and tools that provide ongoing support, especially for the mission-critical components of your IT infrastructure. These activities may be routine, but are critical for meeting service-level commitments and supporting new business initiatives. Use a service manager with pre-engineered and ongoing support
to achieve rapid deployment. A model project plan based on prebuilt components with a project manager ensures smooth deployment. A delivery architect can also be a technical complement to the project manager.
Focused Management
Supporting a dynamic business and resilient IT environment requires effective management of your server
Joseph Gulla is an
executive IT specialist
for IBM data-center services.
His main focus is the monitoring and
management of heterogeneous computing
infrastructures.
Supporting a dynamic business and resilient IT
environment requires effective management of your
server infrastructure and network devices.
Thursday, April 7, 10 am (CST)
REGISTER TODAY: ibmsystemsmag.webex.com, then click on “Event Center.”
WEBINAR
REGISTERTODAY!
German Federal Pension Fund Migrates IBM CICS workloads to Linux on System z Learn how a German government agency simplifi ed operations and met a 350M Euro cost reduction goal by rehosting a mission-critical, 5,000 user application to Linux on IBM System z.
Agenda:Organizations are consolidating workloads onto Linux on IBM System z to lower costs, standardize operations and enhance IT fl exibility. Now you can move online and batch mainframe applications to scalable Integrated Facility for Linux (IFL) processors without sacrifi cing functionality or performance. In this webinar you will learn:
s (OW FOUR DIVISIONS OF THE 'ERMAN PENSION�MANAGEMENT AGENCY�Deutsche Rentenversicherung (DRV), migrated a large, central IBM CICS application from an IBM z/OS environment to Linux on IBM System z
s (OW 5NI+IX -AINFRAME 2EHOSTING SOFTWARE PROTECTS EXISTING INVEST�ments by enabling legacy workloads to run on mainframe Linux partitions
s (OW YOUR BUSINESS CAN BENElT FROM THE BEST PRACTICES AND LESSONSlearned by the DRV and other organizations to maximize benefi ts and reduce risk when replatforming legacy workloads
The DRV intends to use this agile platform to continue consolidation and implement SOA initiatives. Discover how a similar move can benefi t your business.
Who should attend: CIOs and IT managers interested in moving IBM CICS, COBOL, and Batch/JCL workloads to Linux on IBM System z partitions.
ZEN
William Data Systems
ZEN, a z/OS* network-management suite, now features
a unique, automatic SyslogD “system whistleblower” facility. This management suite:¡ Captures, views and notifies any SyslogD events in real-time¡ Alerts users to system events requiring urgent attention the moment they occur¡ Presents multiple source information
on a single, unified console screen¡ Filters alerts enabling the most important to rise above background activity ¡ Communicates alerts to consoles plus smart phones, tablets and laptops ¡ Can automatically trigger further activity via powerful REXX-based automation facilities
OS SUPPORT: All IBM-supported versions of z/OS*
PRICE: Variable URL: www.willdata.com
SolutionsWhat’s new
in the
marketplace
CleverView for TCP/IP v8.2.1
AES
CleverView for TCP/IP v8.2.1 now offers enhanced end-to-end (E2E)
response times for TN3270 session and TN3270 interval System Management File (SMF) records expanding performance monitoring and service level reporting. Highlights of this version include: ¡ TN3270 E2E support that expands the existing application performance
reporting provided by Round Trip Time reports for all TCP/IP transactions¡ MXG support and built-in performance reports that allow CleverView for TCP/IP to produce ETE service level reports for multiarchitecture virtualized environments
OS SUPPORT: z/OS* version 1.4 or higher, z/VM* and Linux* on System z*
PRICE: Variable URL: www.aesclever.com
Vanguard Configuration Manager
Vanguard Integrity Professionals
According to the company, Vanguard Configuration Manager
significantly reduces the cost and time required to test mainframe systems to assess their accordance with the configuration control checklist for z/OS* systems in the National Checklist Program (NCP) of National Institute of Standards and Technology (NIST) and the Department of Homeland Security (DHS). Features include the capability to:¡ Automatically test security configuration controls and verify that a mainframe system is in accordance with the NIST/DHS NCP z/OS checklist (“DISA STIG”) in just a few hours¡ Easily move to continuous monitoring from periodic compliance reporting
OS SUPPORT: z/OS version 1.10 or higher
PRICE: Variable URL: www.go2vanguard.com
i b m s y s t e m s m a g . c o m / m a i n f r a m e M A R C H / A P R I L 2 0 1 1 45
M A R C H / A P R I L 2 0 1 1 i b m s y s t e m s m a g . c o m / m a i n f r a m e46
PlexSpy Application
Status Monitor v1r1
Matter of Fact Software
The enhancements to PlexSpy delivered with version 1.1 can help simplify the
support of CICS* applications and quickly identify the source of issues affecting business systems. Features include:¡ Simplified CICS application management¡ Instant real-time status information concerning applications under investigation
OS SUPPORT: z/OS* 1.10 and higher, CICS Transaction Server V3.1, 3.2 and 4.1
PRICE: Variable URL: www.plexspy.co.uk
Netuitive 5.0
Netuitive Inc.
Netuitive Inc. now integrates with IBM Tivoli*, a leading systems-
management tool for large enterprises. Tivoli customers can now better leverage Netuitive for automated performance
and capacity management in virtualized and cloud environments. Netuitive’s analytics platform:¡ Uses patented behavior-learning technology to replace manual, rules-based approaches¡ Allows enterprises to plug in and synthesize data streams from existing
monitoring sources across silos¡ Forecasts issues before they impact performance and isolates root cause wherever a problem occurs
OS SUPPORT: UNIX* and Windows* PRICE: Variable URL: www.netuitive.com
Predictive Recall
DTS Software Inc.
Predictive Recall is a feature of MONitor that minimizes
production delays triggered by the recalls of migrated data. Migrated production data sets are recalled prior to use, resulting in an overall shorter batch window. To avoid impact on the production workload, the Predictive Recall job can be scheduled to run when the system is underutilized,
prior to the production cycle. This release:¡ Uses the information from the job scheduler to create a list of production data sets that will be accessed in the next production cycle¡ Migrates infrequently used data sets quickly, freeing up DASD space, and recalls them later when they’re needed
OS SUPPORT: z/OS* PRICE: Variable URL: www.dtssoftware.com
Solutions
i b m s y s t e m s m a g . c o m / m a i n f r a m e M A R C H / A P R I L 2 0 1 1 47
For a complete listing of mainframe products, visit our 2011 Buyer’s Guide online: www.ibmsystemsmag.com/mainframe/buyersguide/index.aspx
Advanced Software Products Group, Inc.
www.aspg.com 41, 43
CA
ca.com/chorus 2, 3
Clerity
ibmsystemsmag.webex.com44
ColeSoft Marketing
www.colesoft.com21
Compuware Corp.
MainframeInnovations.com7
DTS Software, Inc.
www.dtssoftware.com23
Edge Information Group, Inc.
www.edge-information.com25
GFS Software
www.gfssoftware.com13
IBM Technical Conferences 39, 47
INNOVATION Data Processing
www.fdr.com/penguins
www.innovationdp.fdr.com/FDRPASdemo
9, CVR 4
Jolly Giant Software Inc.
www.jollygiant.com43
MVS Solutions, Inc.
www.mvssol.comCVR 3
OpenTech Systems, Inc.
www.opentechsystems.comCVR 2
Relational Architects International, Inc.
www.relarc.com5
Trident Services, Inc.
www.triserv.com11
Velocity Software, Inc.
www.velocitysoftware.com1, 15
RESOURCE PAGE
ResourcesAdvertisers’ Index
RESOURCE PAGE RESOURCE PAGE
A Smarter Planet starts with you. Build your skills with IBM Training.
IBM System z Technical Symposium andIBM Power Systems Technical SymposiumApril 27-29 | Orlando, FL - Attend two events for the price of one!
Is Spring the best time for you to update your skills? By now you have heard all the buzz on the strategic IBM zEnterprise™ and IBM POWER7 Systems™. Here is your chance to get the latest details, as well as, update your skills on current systems.
IBM System z Technical Symposium
featuring a tecnical update on zEnterprise, z/OS®,
z/VM®, z/VSE® and Linux on IBM System z®.
IBM Power Systems Technical Symposium
featuring a technical update on IBM Power
Systems™ AIX®, Linux® and IBM i.
AIX, IBM, the IBM logo, ibm.com, Power Systems, POWER7 Systems, Smarter Planet, System z, zEnterprise are trademarks or registered trademarks of Interna-tional Business Machines Corporation in the United States, other countries, or both. A current list of IBM trademarks is available on the Web at “Copyright and trademark information” at: ibm.com/legal/copytrade.shtml. Linux is a registered trademark of Linus Torvalds in the United States, other countries, or both.
Enroll today! Early bird pricing available until March 27th.
ibm.com/systems/conferenceseriestwitter.com/IBMTechConfs
Sponsor and Exhibitor opportunities available. Visit ibm.com/training/solutioncenter
Bijan Davari, who pioneered critical functions of the modern computer chip, was born that same year in Tehran. At age 11 his dad handed him his first transistor radio. “It was so much smaller and more elegant,” says Davari, who tinkered with vacuum-tube radios. “I could see that it was a game changer, and that’s how I got started.”
By the time he was in his 30s, Davari was leading the IBM research team that would produce another game changer—the first generation of high-performance, low-voltage logic (complementary metal oxide semiconductor, or CMOS) that characterizes computer chips. “I’ve always been interested in the ‘how’ of making significant advances in function, particularly when it’s believed it can’t be done,” Davari says. Today we take it for granted that these chips, which have gone from megahertz to gigahertz
performance standards within 15 years, power everything from iPhones to supercomputers. Without Davari’s breakthroughs, computers would be much hotter, slower and noisier.
Right from the start, Davari and his team challenged conventional thinking. “Assumptions were that you had to have high voltage to get a lot of power,” he says. “If you lowered voltage, it was assumed performance would get worse.” No one wanted to bother much with CMOS, a low-voltage technology that powered wristwatches. Davari’s team proved if you lowered the voltage, scaled down chip components and used different materials, CMOS became powerful. “You’re driving less load, so you don’t need as much current,” he adds. “If you’re driving large loads to make something fast, they get hot, and you get to a point where you can’t cool it.”
He didn’t stop there. In 1996 Davari was named an IBM Fellow—the company’s highest honor—for his work; last year he received an award from IEEE (Institute of Electrical and Electronics Engineers) for outstanding contributions to his field.
Today Davari is working to achieve significant technological breakthroughs at the system level as vice president of Next Generation Computing Systems/Technology at IBM. Future performance gains will come through more integration—such as IBM’s new electrical/optical device chip—as well as through parallel processing at the system and software levels. Davari says we’ll eventually be able to connect to very powerful, highly specialized applications through cloud and wireless environments.
“You’ll be able to participate in a professional soccer game, for example, without changing the outcome,” he says. “The distinction between the real and computing world will be totally blurred.”
Game Changer Bijan Davari changed the computer chip By Sara Aase
In 1954, Texas Instruments engineers radically minia-turized the transistor radio, making it simultaneously less expensive and more powerful. Their reimagination of what
had been a slow-selling semiconductor technology changed the world: the transistor radio became the most popular electronic device in history, and the semiconductor, of course, became the heart of modern technology.
M A R C H / A P R I L 2 0 1 1 i b m s y s t e m s m a g . c o m / m a i n f r a m e48
Stop RunAn eclectic
take on the
mainframe
world
Sara Aase is a Minneapolis-
based freelance writer.
Bijan Davari is vice
president of Next
Generation Computing
Systems/Technology
at IBM.
For AFFORDABLE PRICING and a FREE 90-Day Trialcall 973.890.7300, e-mail [email protected] or visit www.fdr.com/penguins
CORPORATE HEADQUARTERS: ����!'/(-.,+��0(���)//*(�'**.����� �����2��������� ��� �2�'1���������� �����E-mail: [email protected] 2�[email protected] 2�http:/ / www.innovationdp.fdr.com
EUROPEAN FRANCE GERMANY NETHERLANDS UNITED KINGDOM NORDIC COUNTRIESOFFICES: ����������� � ������� �� ���������� � ��� ������ ��������������
FDR /UPSTREAM uses the tools and systems that you have been using for decades;
integrating with your existing z/OS tape management, security and scheduling systems.
FDR/UPSTREAM Linux on System z
If your organization is like many others, you are finding that the values and benefits of running Linux on System z
are plentiful. Our customers report that server consolidation and moving those open systems applications onto
Linux on System z results in �Lower cost of operations �Less servers �Fewer software licenses �Fewerresources to manage �The mainframe legendary dependability �Less energy, cooling and space.
And given that the Linux data is now on the System z mainframe who better than INNOVATION,
the makers of FDR, to provide the backup tool!
FDR/UPSTREAM provides reliable, file level data protection by backing up Linux on System z data directly
to z/OS mainframe tape or DASD. UPSTREAM backups can be automatically initiated from the z/OS mainframe
and provide high speed operations including support for HiperSockets, with data de-duplication features and
online database agents. UPSTREAM operations can be completely automated with JCL condition code
checking, automated email notification, and exception reporting, combining the data protection of Linux on
System z data into existing disaster plans.
VISIT US AT: SHARE TECHNOLOGY EXCHANGE 4 BOOTH #300 4 FEB 28 - MAR 2, 2011 4 ANAHEIM, CA
View the FDRPAS Product Demo…www.innovationdp.fdr.com/FDRPASdemo
CORPORATE HEADQUARTERS: ����!'/(-.,+��0(���)//*(�'**.����� �����2��������� ��� �2�'1���������� �����E-mail: [email protected] 2�[email protected] 2�http:/ / www.innovationdp.fdr.com
EUROPEAN FRANCE GERMANY NETHERLANDS UNITED KINGDOM NORDIC COUNTRIESOFFICES: ����������� ��������� ������� �� �������� ���������� ������������ � ��� ��������������������������������
A visual and audio overview of our product line.Now available for: FDRPAS, FDRMOVE, FDRERASE for z/OS, FDRERASE/OPEN and FATSCOPY.
For a No-Obligation FREE Trial and a FREE Keylight call: 973-890-7300, e-mail: [email protected] or visit: www.innovationdp.fdr.com
FDRPAS...Simplified Automated MigrationFDRPAS, is a DASD migration facility that is fast, safe, and non-disruptive providing
continuous availability. With FDRPAS you can move individual disk volumes, or an entire
DASD controller, to new hardware during normal system operations, without interrupting
any system activity.
8 Generate all FDRPAS jobs from ISPF panels
8 Comprehensive Simulation features that automatically determine any issues
that need to be addressed like problems in the VVDS and the VTOCs
8 Vendor independent…supports any manufacturers’ disks
Competitively priced with flexible short-term rental periods available. FDRPAS over1,400 migrations since 2001 with a proven record of performance and reliability.
Top Related