CDMP Study Group

SESSION 5 – DATA ARCHITECTURE

April 1, 2020
Laura Sebastian-Coleman, Ph.D., CDMP
Advisor to the DAMA New England BOD
Email: lsebastiancoleman@gmail.com
AGENDA
• Facilitator Introduction
• Chapter 4: Data Architecture
• Overview
• Introductory Note
• What data architecture is, how it fits with other architectural domains, and why it is important
• Architectural frameworks and why they are important
• Architectural artifacts: Enterprise Data Model and Data Flow Diagrams
• Approach to studying
•Q&A
• Next Session

New England Data Management Community

Facilitator
Laura Sebastian-Coleman, Ph.D., CDMP
 DQ Lead, Finance DG COE, Aetna/CVS Health
 DAMA New England Board Advisor
 Production Editor, DMBOK2
 Author:
 Navigating the Labyrinth
 Measuring Data Quality for Ongoing Improvement

CONTACT INFO:
EMAIL: lsebastiancoleman@gmail.com
PHONE: 860-983-0399

New England Data Management Community

INTRODUCTORY NOTE

This study group is offered as a service of DAMA New England for DAMA New England
members. It not an official, DAMA International authorized training course because DAMA-I has
not yet created an authorized trainer program.

The purpose of this group is to help prepare members to take the CDMP. We will do so by
reviewing the content of chapters of the DMBOK2.

DAMA New England makes no claims for the effectiveness of the sessions or the ability of
participants to pass the CDMP exam after having attended. In fact, you should plan on doing a
lot of individual study to pass the exam.

All tables and images, unless otherwise noted, are copyrighted by DAMA or based on DAMA-
DMBOK2 tables and images.

New England Data Management Community

Thinking about Chapter 4: Data Architecture
• No organization starts from scratch, from a data perspective. Even if
one did, data dependencies arise quickly.

• Architecture enables a BIG PICTURE view of an organization and its

data. Architectural artifacts are the means by which an organization
represents its data to itself.

• It also enables a LONG-TERM view of an organization and its data.

Organizations need an architectural ROADMAP to understand where
they are, as well as to understand where they want to be and how to
get there.

• An architectural perspective is critical to how an organization

establishes its data strategy. Architecture can DRIVE INNOVATION.

• If we recognize, as Tom Redman says, that DATA IS DONG ITS WORK

when it is moving through the organization and BEING USED, we will
see why ideally, data architecture practices should be comprehensive,
strategic, and dynamic.

New England Data Management Community

Chapter 4: Data Architecture
Identifying the data needs of the enterprise and designing and maintaining the master blueprints to meet those
needs. Using master blueprints to guide data integration, control data assets, and align data investments with
business strategy.

The word Architecture can refer to:

• A description of the current state of systems
• The components of a set of systems
• The discipline of designing systems (architecture practice)
• The intentional design of a system or a set of systems (future state or
proposed architecture)
• The artifacts that describe a system (architecture documentation)
• The team that does the design work (the Architects or the Architecture
team)

New England Data Management Community

Data Architecture: Part of Enterprise Architecture
• Enterprise Architecture encompasses: • Successful Architecture depends on
• Business architecture • Good design
• Data architecture • Planning
• Application architecture • Ensuring that the designs and plans are
• Technology architecture executed effectively

• Enterprise architecture enables an • Data Architects seek to

organization to: • Design an optimal technical footprint
• Understand the current state of their systems • Improve operational and project efficiencies
• Promote desirable change toward future • Increase the ability of the organization to use
state its data.
• Enable regulatory compliance • Data Architecture is most valuable
• Manage data when it supports the needs of the
• Manage the systems in which data is stored enterprise.
and used

New England Data Management Community

Data Architecture: Part of Enterprise Architecture
Domain Enterprise Business Enterprise Data Architecture Enterprise Applications Enterprise Technology
Architecture Architecture Architecture

Purpose To identify how an To describe how data should To describe the structure and To describe the physical
enterprise creates value be organized and managed functionality of applications technology needed to
for customers and other in an enterprise enable systems to
stakeholders function and deliver
value

Elements Business models, Data models, data Business systems, software Technical platforms,
processes, capabilities, definitions, data mapping packages, databases networks, security,
services, events, specifications, data flows, integration tools
strategies, vocabulary structured data APIs

Dependencies Establishes Manages data created and Acts on specified data Hosts and executes the
requirements for the required by business according to business application architecture
other domains architecture requirements
Roles Business architects and Data architects and Applications architects Infrastructure architects
analysts, business data modelers, data stewards
© DAMA-International, 2017
stewards

New England Data Management Community

Data Architecture: Business Drivers

The problems Data Architecture is trying to Data architecture enables organizations to:
solve: • Plan for evolution: Prepare organizations to
• More data than the average bear! Most evolve products, services, and data to take
organizations have more data than a single advantage of business opportunities inherent
person can understand. in emerging technologies
• Different Levels Of detail for different • Translate requirements: Translate business
decision-makers. Data needs to be needs into data and system requirements so
represented at different levels of detail so that processes consistently have the data they
that people can make decisions about it. require
• Architectural Artifacts help with • Manage delivery: Manage complex data and
comprehension. Data architects create the information delivery throughout the
artifacts that enable people to understand enterprise
their organization’s data. • Facilitate alignment between Business and IT
• Act as agents for change, transformation, and
agility

New England Data Management Community

Data Architecture from different perspectives
• Chapter Four looks at Data • Data Architecture artifacts (master
Architecture from three blueprints) are used to:
perspectives: • Define data requirements
• Guide data integration
• Data Architecture outcomes / • Control data assets
artifacts: models, definitions and data • Align data investments with business
flows on various levels of abstraction strategy
• Data Architecture activities, to form, • Data Architects must
deploy and fulfill Data Architecture
intentions • Collaborate with, learn from and
influence various stakeholders that
• Data Architecture behavior: are engaged with improving the
collaborations, mindsets, and skills business or IT systems development
among the various roles that affect
the enterprise’s Data Architecture. • Help to establish the semantics of an
enterprise, via a common business
vocabulary

New England Data Management Community

Architectural frameworks are
important to understanding
architecture as a whole.

Of these, the most influential

is the Zachman Framework,
which is organized around
basic questions related to the
perspectives of stakeholders
across an enterprise.

Different stakeholders require

information at different levels
of abstraction.

This is wicked hard to read, so

the next slide has the
simplified version that appears
in DMBOK2.
 What How Where Who When Why

Inventory Process Distribution Responsibility Timing Motivation

Executive Identification Identification Identification Identification Identification Identification Scope Context

Inventory Process Distribution Responsibility Timing Motivation

Business definition Definition Definition Definition Definition Definition Business
Management Concepts

Inventory Process Distribution Responsibility Timing Motivation

Architect Representation Representation Representation Representation Representation Representation System Logic

Inventory Process Distribution Responsibility Timing Motivation

Technology
Engineer Specification Specification Specification Specification Specification Specification
Physics

Inventory Process Distribution Responsibility Timing Motivation Tool

Technician Configuration Configuration Configuration Configuration Configuration Configuration Components

Inventory Process Distribution Responsibility Timing Motivation Operational

Enterprise Instantiations Instantiations Instantiations Instantiations Instantiations Instantiations Instances

Inventory Process Distribution Responsibility Timing Motivation © DAMA-International, 2017

Sets Flows Networks Assignments Cycles Intentions
Data Architecture: EDM and Data Flow Design
Data exists in and moves through an enterprise.
Future State Enterprise Future State
Data Architecture needs to account for both the current and Data Model Data Flows
future state of the data model and the data flows.
Enterprise Data
Enterprise Data Model (EDM): The EDM is a holistic, enterprise-level, Architecture
implementation-independent conceptual or logical data model providing a
common consistent view of data across the enterprise. It sets forth the
Current state Enterprise Current State
foundation for all data and data-related projects. … Any project-level data
Data Model (EDM) Data Flows
model must be based on the EDM. The EDM should be reviewed by
stakeholders, so that there is consensus that it effectively represents the
enterprise.

Data Flow Design: Defines the requirements and master blueprint for
storage and processing across databases, applications, platforms, and
networks (the components). These data flows map the movement of data
to business processes, locations, business roles, and to technical
components.
New England Data Management Community
The EDM Enterprise Data Model
1 diagram:
12-20 significant
Conceptual Model business subject
areas with
The EDM can be a stand-alone artifact, or it can be comprised of relationships
the overall set of models created by the enterprise.

Different types of models – at different levels of abstraction –

Subject Subject Subject Subject 1+ diagram per
must be related to each other. subject area:
Area Area Area Area
50+ significant
Model Model Model Model entities within
Conceptual models are ultimately linkable to physical subject area with
relationships
application data models

The diagram shows: For each Subject

Area Logical
• A conceptual overview over the enterprise’s subject areas Model:
• Views of entities and relationships for each subject area Logical Logical Logical Logical Increase detail by
adding attributes,
• Detailed, partially attributed logical views of these same Model Model Model Model and less-significant
entities and
subject areas relationships
• Logical and physical models specific to an application or
Limit scope to the
project subject and 1-step
LDM LDM LDM LDM LDM LDM LDM
external
relationships
PDM PDM PDM PDM PDM PDM PDM Limit scope to
the subject physical
objects and
Application or Project-Specific relationships
New England Data Management Community © DAMA-International, 2017
The Subject Area Model
As stated, the EDM can be a stand-alone artifact, or it can be Product Design Commercial Sales Subject
Subject Area Offer Subject Area
comprised of the overall set of models created by the Area
enterprise. Market
Product Sales
Product Group Offering
Platform Order
Portfolio
Establish a principle for division of subject areas (a subject area
discriminator). For example, Offering Range Occurs in
Uses
• Normalization rules
• Systems portfolios (i.e., funding)
• Data governance structure and data ownership
Sales Sales
(organizational) Belongs to Product
Item Order Item
• Top-level processes (based on the business value chains)
• Business capabilities (enterprise architecture-based)
Product Design
Bill-of-material Specifies
(BOM)
The Subject Area structure is usually most effective for Data Sales Bundle
Bill-of-material
(BOM)
Architecture work if it is formed using normalization rules.
Product
Part Sales
Details Order Item
Part Configuration
Structure © DAMA-International, 2017

New England Data Management Community

Data Flows Service & repair statistics © DAMA-International, 2017
Material returns

Data Flow diagrams depict how data

CRM
moves through an organization. Data
flows can be organized in different Service
ways. Customer data & repair
Customer statistics
agreement

This traditional high-level flow

diagram shows the movement of data
between different systems that Product Design BOM
Product properties Aftermarket
support business processes. PDM System Sales System
System

Product design BOM

Such a diagram can be created with Product properties Sales BOM
Order Contents
Sourcing data
different levels of detail, depending
on the audience and purpose.
Manufacturing Manufacturing
Routing System Planning Serial numbers
Manufacturing BOM System Individual product BOM
Production line routing
Lead times

New England Data Management Community

Data Flows
Business Processes
This matrix shows the relationship between
Product Marketing Industrial Order
Manufacturing Logistics Invoicing
business processes and data entities that development & Sales preparation management

create or use data. Product

Product Part
Benefits of a matrix depiction: Manufacturing
Plant
• Shows that data does not flow in just one Customer
direction

Major Entities
Sales Item
• Shows many-to-many relationships Assembly
Structure
• Can clarify data acquisition responsibilities Sales Order
• Can clarify data dependencies between Production
Order
processes Individual
Product
Shipping
Customer’s
Invoice

Read/use © DAMA-International, 2017

Create

New England Data Management Community

Data Architecture Activities: Establish the practice
Establish Data Architecture Practice
Relationship to projects
• Part of Enterprise Architecture OR adopt a
• Defining requirements
framework
• Reviewing project data design
• Determining data lineage impact
Account for
• Data replication control
• Strategy
• Enforce Architecture standards
• Culture
• Guide technology decisions
• Organizational accountabilities,
responsibilities
Evaluate Existing Specifications
• Work methods
• Results / relation to roadmap
Develop a roadmap

Manage enterprise requirements within

projects

New England Data Management Community

Data Architecture Activities
Evaluate and Update Existing Specifications
Manage enterprise requirements within
projects
Develop a roadmap
• Bring an enterprise perspective to project
• Manage dependencies
scope
• Make forward-looking decisions
• Understand requirements
• Evaluate trade-offs
• Ensure implementation makes sense in
• Formulate pragmatic plans, aligned with
terms of the overall architecture
business needs and opportunities, external
requirements, and available resources.
Note: The role of the enterprise data architect
• To develop a roadmap, start with the
will depend in part on the implementation
lowest dependency activities
methodology (e.g., waterfall, agile, hybrid0

New England Data Management Community

Discussion / Q&A: How to study …
Architectural Frameworks

Architectural Artifacts

The role of the Data Architect

Strategy / Projects
Innovation / Quality

New England Data Management Community

NEXT SESSION
Date Topic Facilitator
February 19th Chapter 1: Data Management Tony Mazzarella
March 4th Chapter 2: Data Handling Ethics Lynn Noel
March 18th Chapter 3: Data Governance Sandi Perillo-Simmons
April 1st Chapter 4: Data Architecture Laura Sebastian Coleman
April 15th Chapter 5: Data Modeling & Design Lynn Noel
April 29th Chapter 6: Data Storage & Operations Karen Sheridan
May 13th Chapter 7: Data Security Laura Sebastian-Coleman
May 27th Chapter 8: Data Integration & Interoperability Mary Early
June 10th Chapter 9: Document & Content Management Sandi Perillo-Simmons
June 24th Chapter 10: Reference & Master Data Mary Early
July 8th Chapter 11: Data Warehousing & Business Intelligence Tony Mazzarella
July 22nd Chapter 12: Metadata Management Karen Sheridan
August 5th Chapter 13: Data Quality Laura Sebastian-Coleman
August 19th Chapter 14: Big Data & Data Science Nupur Gandhi
September 2nd Chapter 15: Data Management Maturity Assessment Laura Sebastian-Coleman
September 16th Chapter 16: Data Management Organization & Role Expectations Agnes Vega
September 30th Chapter 17: Data Management & Organizational Change Management Tony Mazzarella
October 7th Final Review Tony Mazzarella

New England Data Management Community

HOMEWORK – Data Modeling

What are the primary benefits of data modeling to the

organization?
What behaviors help organizations realize these
benefits?
How can they get more value from data modeling?

New England Data Management Community