News
 Distributed
Data Management: A Technology View
By Raj Nathan, senior vice president & CMO for
Worldwide Marketing & Business Solutions
Operations for Sybase
By now, IT and business staff alike understand that
data is an organization’s most valuable asset.
They’ve come to realize that data has a value beyond
the application through which it was created and
beyond the silo in which it resides. Accurate,
consistent, current data is a strategic asset that
has the power to increase competitive advantage,
drive revenue growth and market share, and enhance
an organization’s reputation. The key to unleashing
its value is to make it available anywhere, anytime
to the individuals and groups that can put it to
work in the service of the organization’s mission
and strategies. Distributed data management is
essential to driving the enterprise mobility
strategies that increase productivity, reduce
expenses, and create efficiencies in a tight
economic environment and ultra-competitive
situations.
When discussing distributed data management, it is
important to understand that in all likelihood a
great deal of data distribution has already occurred
in your organization. Data always seems to find a
way out of application- and department-specific
silos, databases, operational systems, data
warehouses and other sources from mainframes to
smartphones.
It is also important to take stock of the
individuals and groups of people in your
organization who create, collect, manage, move,
stage, update, define and ultimately use data to
make business decisions. These people go by many
names. In IT departments they include application
developers, database administrators, storage system
administrators, network administrators,
infrastructure architects, and data architects. On
the business side of the organization, there are
additional data creators, consumers, and other data
stakeholders. These individuals also go by many
names, including line-of-business managers,
salespeople, service personnel, business
strategists, planners and analysts, compliance
officers, and more.
Creating and implementing a distributed data
management system that provides users with accurate,
consistent, auditable, secure and timely information
with which to perform their jobs has at least one
other hurdle to overcome.
Customer? Just what do you mean by that?
Not all of the people touching and manipulating data
speak the same language. Individuals and departments
define common data elements differently, creating
confusion across the enterprise. An example with
which you may be familiar is when a senior manager
asks for a list of customers. Sounds simple enough,
right?
The problem is, however, that he gets different
reports depending on the particular department or
system he consults. The organization’s accounting
system generates a report containing 500,000
customers. But the customer relationship management
system indicates that there are more than a million
customer IDs. That’s quite a discrepancy, but it all
comes down to how the attribute “customer” is
defined in each system. And while it’s
understandable, given that it’s not unusual for an
organization to have a dozen or more points of entry
for customer information, the result is that getting
something as seemingly simple as a consolidated,
reliable view of “customers” can be frustrating -
even maddening. To reconcile such discrepancies, a
senior manager may have to assign a small army of IT
staff to sort through the data to understand the
different categorizations of data attributes and
de-duplicate data from multiple applications and
systems in order to arrive at a reliable answer.
This, of course, can take time, which limits the
organization’s agility and competitiveness.
So, moving to distributed data management is not as
simple as taking a technology wrecking ball to
traditional silos and moving the data to the people
who need it, where and when they need it, and in the
formats in which they need it. (Although that would
be challenging enough.)
Where to Begin
Whether your organization is brand new and building
a database management system afresh, or more
typically, an existing organization struggling to
wrap its arms around enterprise data residing in
multiple locations, in multiple systems on multiple
platforms in order to gain a holistic view of
customers, products, or overall company performance,
it is critical to begin by integrating your data
before you begin distributing it around your
enterprise.
As enterprise data environments have become
increasingly diverse and complex, data professionals
are turning to modeling tools to help them do that
in an effective and efficient way. Using a data
modeling tool, data and enterprise architects can
create and capture the metadata needed to describe
their data environments. They can establish
requirements, create data models, data flows,
process models, and more that enable them to create
and maintain a single version of the truth, from
which everyone in the organization can work with
confidence.
In organizations in which data resides in a plethora
of departmental silos, application-specific
databases and other locations, architects can use
modeling to reverse-engineer existing data flows and
then model and implement changes to those flows as
required by new business processes. These
capabilities allow the business to anticipate change
more proactively and make rapid adjustments
centrally as needed. Again, the goal is to ensure
that across the enterprise, everyone is working from
the same data. This is the essence of what is meant
by the phrase “aligning IT and the business.”
Data modeling is extremely powerful and effective.
But additional prework and processes should be put
in place before deploying the technologies that will
enable you to distribute data around your
enterprise.
Data Governance and Master Data Management
Since enterprise data is such a strategic asset and
so critical to the efficient functioning and growth
of organizations, it must be subject to business
rules and consistent definitions in order to deliver
optimal value.
Data governance is a discipline and management
category intended to improve the quality and
accuracy of enterprise data. It does this through
the creation and enforcement of standard data
definitions, the designation of roles and
responsibilities for the stewardship of data to
ensure that enterprise data - regardless of its
origin - is available, accessible, reliable,
consistent, auditable, and secure.
Data governance is also the prerequisite to and
enabler for master data management, which puts those
business rules and definitions to work to ensure
that there is consensus across the organization on
how data is classified, how it is to be integrated,
accessed, created, updated, monitored, and by whom.
Now You’re Ready for the Technology Discussion
and Deployment
Only once you’ve got a handle on the
business-critical data residing across your
enterprise-data that is essential to understanding
and advancing your business-are you ready to talk
about technology solutions.
Yes, distributing data across the enterprise to
various constituents with different requirements
implies a number of technology solutions. These
include high-performance transactional and
analytical database management systems, replication
technology, robust, high-speed networks, layered
security systems, disaster recovery and business
continuity systems, and more.
Once you’ve broken down silos and decoupled data
from applications, you may also find that you need
to create a centralized location from which users
can access master reference data, according to their
requirements and permissions.
One approach that works well, particularly in
organizations that have many silos and legacy
systems, is to create a data services layer that
becomes the source of the data against which users
can run queries, and from which role-specific data
can be pushed out to users on a scheduled or
subscription basis. Such a layer, which spans
multiple systems to bring together reliable,
integrated, master data, can function as the heart
of an efficient, maintainable, distributed data
management system, while shielding users from the
underlying complexities and preventing a
reoccurrence of the confusion regarding questions
like “How many customers do we have?”
Many companies have employed the approaches
described in this article and have achieved
significant results, including increased
efficiencies, improved views of business
performance, enhanced competitive advantage,
simplified regulatory compliance, and the peace of
mind of knowing that business decisions are being
made based on accurate, holistic information.
---Source:
Information Management Magazine June 4, 2009 (www.information-management.com).
Raj Nathan is the senior vice president and chief
marketing officer for Worldwide Marketing and
Business Solutions Operations for Sybase and its
subsidiaries, Sybase iAnywhere and Sybase 365.
|
|
|
Melissa Data
|
 |
| Enhance your
website, software or database with
easy-to-integrate data quality programming tools
and web services. |
|
|
|
|
 |
|
Save money on postage using leading
mail preparation software and other
direct marketing products. |
|
|
|
|
 |
Update & standardize addresses and
find out more about contacts in your
database.
|
|
|
|
|
 |
Find new customers perfect for your
business with our online and
specialty mailing lists.
|
|
|
|
|
 |
Locate the business information you
need such as ZIP Codes, address
verification, maps.
|
|
|
|
|
Facebook
Twitter
