Data Governance

43

3

Metadata Management

and Data Governance

Michael Schrader

CONTENTS

What Is Governance? ...................................................................................... 44

Why Governance? ........................................................................................... 44

Fit for Purpose versus Single Version of the Truth ......................................45

Governance Board ............................................................................................47

Program versus Project Governance .............................................................47

Metadata Framework Dened ........................................................................47

Metadata Types ................................................................................................ 48

Business Metadata ...................................................................................... 48

Searchable ................................................................................................49

Easy to Access and Integrated ...............................................................50

Exposed ....................................................................................................50

Technical Metadata......................................................................................50

Process Metadata .........................................................................................51

Data Lineage ............................................................................................52

Impact Analysis .......................................................................................52

Operational Metadata .................................................................................53

Case Study of Metadata Use for Data Governance ......................................53

Master Data Management .............................................................................. 54

Metadata Repository Types ............................................................................. 55

Data Proling: Investigation and Enforcement ........................................... 56

Data Life Cycle ..................................................................................................58

Data Retention Techniques .............................................................................59

Data Versioning ............................................................................................... 60

Data Versioning Techniques ..................................................................... 60

Star Dimensional Model Data Versioning .......................................... 60

ird Normal Form Data Versioning ..................................................63

44  •  Michael Schrader

WHAT IS GOVERNANCE?

Data governance is the process of creating and enforcing standards and

policies concerning data. Data governance standards and policies are

created and enforced by a single authoritative organization called the

Governance Board.

e governance process isn’t a transient, short-term project. e gover-

nance process is a continuing enterprise-focused program.

Governance provides standards and policies around the following in

relation to processes and data:

• Soware products

• Infrastructure

• Quality

• Security

• Dispute resolution

• Life cycle

• Best practices

• Architecture and future road maps

• Project prioritization

• Asset management

• Version control

• Evangelizing and communication

• Vendor relationship management

• Legal and corporate compliance

WHY GOVERNANCE?

A strong governance program is vital to the success of any enterprise

architecture. It provides:

• Compliance: Governance programs allow for the compliance to reg-

ulatory requirements. We have all heard “I am too pretty to go to

Enforcement ......................................................................................................63

Conclusion ....................................................................................................... 64

Metadata Management and Data Governance • 45

jail.” Well, without governance, it’s true; there exists no formalized

process for proving for regulatory compliance with HIPAA (Health

Insurance Portability and Accountability Act, 1996) and privacy

laws. Data governance initiatives may be aimed at achieving a num-

ber of objectives including oering better visibility to internal and

external customers and compliance with regulatory laws.

• Harmonizing: Governance provides for standard denitions. is

allows developers, database administrations, end users, and data

stewards to be working on the same page.

• Consistent Analysis: is allows the business to roll up (consolidated)

values with consistent values comparing apples to apples.

• Faster Development: By providing standard denitions and models,

we provide the infrastructure for extreme development. e most

dicult part of the development of a BI (business intelligence) proj-

ect is getting the data. If we do not need to add data elements, we can

develop applications in days or hours.

• Coniction Resolution: Many organizations have evolved into iso-

lated fathoms. We need an authority to resolve disputes across each

of the efdoms.

• Asset Management: is is the harvesting and management of assets

to maximum business returns. Asset management ensures prioriti-

zation with an enterprise view and allows reduction in costs through

elimination of duplicate eorts.

• Security: Operational metadata. Who accesses what and when and

how? is will not only allow for regulatory compliance, but for better

data warehouse design. We can map the usage of each data mart.

Ifone data mart is not being utilized, we can redesign or start an

educational program on how to better utilize it.

• Better Data and Process Quality: By having clearly dened enter-

prise business rules all users will be able to understand. We can

further empower end users through exposing them to the business

rules. is allows for completing jobs at the speed of business.

FIT FOR PURPOSE VERSUS SINGLE VERSION OF THE TRUTH

In a Single Version of the Truth architecture, the Governance Boards

agree upon one true denition of each attribute. Without a governance