KCDC2015 - Enterprise Architecture In A Heterogeneous Environment

Enterprise Architecture in a Heterogeneous EnvironmentBuilding software landscapes in the real

world

Dustin Hudson2502A| 2:30p

Dustin Hudson Enterprise Architect/Manager @ Seaboard Foods10+ years in the field

• Visual arts• Graphic design

• Visual marketing• Web marketing• Development

• Business intelligence• Systems architecture

• Enterprise architecture

Titanium Sponsors

Platinum Sponsors

Gold Sponsors

Let’s start with a definition• Enterprise architecture (EA) is a discipline for

proactively and holistically leading enterprise responses to disruptive forces by identifying and analyzing the execution of change toward desired business vision and outcomes. EA delivers value by presenting business and IT leaders with signature-ready recommendations for adjusting policies and projects to achieve target business outcomes that capitalize on relevant business disruptions. EA is used to steer decision making toward the evolution of the future state architecture.

--Gartner

And in real world terms• Enterprise architecture is about strategic technical

direction. • Let’s unpack that phrase: • Strategic – big-picture, long-term, visionary• Technical – technology-oriented, IT-driven• Direction – future-thinking, moving forward

• Simply put, the EA is the person who draws the map from where we are to where we want to be.

Cartography as an art formTo draw a map, we need to know three things:

Where are we?What do we have in our portfolio? What are our current business processes?

Where do we want to go?What do we want to change in our portfolio? What improvements do we want to make in our business processes?

How do we get there? What decisions should we make now to move towards our goals? What decisions should we not make because they detract from our goals?

“A good plan is like a road map: it shows the final destination and usually the best way to get there.”

--H. Stanley Judd

Ivory tower architecture• Carefully controlled• Intricately planned• All technologies perfectly

aligned• All applications align to the

model• Continuous improvement• Short system lifecycles• Strong gate check

Meanwhile, in the real world…• Broad mix of technologies• Some built, some bought,

some bought-but-customized-to-hell• Business decisions that

ignore architecture concerns• Legacy technical decisions

still haunting the landscape• System lifecycles that

measure in geological eras

Ok, so…what can we do about it?

“In theory, there’s no difference between theory and practice. In practice, there is.”

--Jan ve de Snepscheut (attributed)

• Deep breath• Acknowledge it won’t ever be

perfect• Our job is to keep getting

closer

Big picture stuff first• First, do no harm• Any solution should meet three criteria, in this order: • Effective – does it solve the problem?• Efficient – does it solve the problem in a way that makes life

easier?• Elegant – does it solve the problem as simply as possible?

• Remember our definition of EA – strategic technical direction• Every solution should further the overall strategy• Future-proof wherever possible• Keep moving forward

Now, deal with reality• Identify the problem to be solved• Acknowledge what cannot be changed• Build a plan that incorporates those things• Use what you can control to mitigate what you cannot

• Fix the problem and leave the world a better place• Look for opportunities to clear debt and build foundation

• Communicate the what and the why• Add in the how when necessary, but generally let the guys

in the weeds handle that• Visualize – a picture is worth a thousand meetings

Some specific ideas• Problem statement• Goals statement• Business goals• IT goals• Shared goals

• Core principles• SWOT analysis• Road maps• As is/to be• Prioritization

• Flow diagrams• Data• Process• User

• Milestone diagrams• Problem area• Core operational model• Relationships

Real world example #1• Problem: After years of growth, it’s time for an enterprise-

wide systems modernization effort to move away from outdated technologies. • Goals:

• Implement the core applications in modern technologies• Build a foundation for future growth, allowing for a shifting

system landscape

• Approach: • Build a standard integration foundation using ETL, modeled after

SOA patterns• Isolate applications from each other using an endpoint model for

integrations

Real world example #1• Each application lives

in its own little world and interacts via endpoints• Single exception is for

passive reads direct to EDW

• Supports integrated applications/reporting but still isolates applications from unrelated changes

Application A

Application B

Endpoint

Endpoint

Application CEndpointStaging

EDW

BI

Application DEndpoint

Real world example #2• Problem: The business model is highly integrated, but the

data is isolated by system of record. Business areas define terms independently. Business intelligence usage is growing but fragmented. • Goals:

• Create a single source of the truth across business areas. • Support data models that reflect the integrated nature of the

business. • Build a foundation for future business intelligence growth.

• Approach: • Implement an enterprise data warehouse as a common repository. • Leverage the systems integration landscape to support the EDW.

Real world example #2• EDW is the primary

repository for all business intelligence reporting. • Multiple systems

provide data to the EDW. • EDW can leverage data

already flowing through the integration frameworks to populate.

Application A

Application B

Endpoint

Endpoint

Application CEndpointStaging

EDW

BI

Application DEndpoint

Real world example #3• Problem: A heterogeneous systems landscape has resulted in non-standard

change deployment processes. Different technologies and systems deploy changes in different, sometimes contradictory ways.

• Goals: • Achieve a common process flow for deploying changes across environments while still

accounting for the disparate mechanisms required to implement changes• Enforce a separation of duties for security and auditing purposes • Align with change management efforts originated from the PMO

• Approach: • Identify a vertical environment chain wherever possible• Implement a security matrix identifying who is responsible for making and deploying

changes by type and environment• Align with change management processes by providing an authorization step in the

process flow• Define high level change deployment actions, but acknowledge that specific tasks are

mandated by the technical stack in question

Real world example #2• Process is

standardized at the task level but adaptable at the action level. • Separation of duties is

enforced.• Change type and

environment determines role

• PMO authorization is required to move above Dev• Enables change

management controls

Local Development Test Production

1 2 5 8

4 7

Authorization

Execution

3 6

Task ID Title Description Owner Environment1 Change is Executed Requested change is executed by the developer Developer Local2 Change is Deployed to Dev Change is deployed from Local to Dev environment Developer Development3 Change is Internally Tested Change goes through internal QA process Developer Development4 Change is Authorized for Test Change is confirmed ready for deployment to Test PMO Development5 Change is Deployed to Test Change is deployed from Dev to Test environment Administrator Test6 Change is Externally Tested Change goes through user acceptance testing Administrator Test7 Change is Authorized for Prod Change is confirmed ready for deployment to Production PMO Test8 Change is Deployed to Prod Change is deployed from Test to Production environment Administrator Production

Thank you.

Dustin Hudson

Seaboard Foods

[email protected]

www.linkedin.com/in/dustinhudsonkc

References & Reading List:

• Enterprise Architecture as Strategy: Creating a Foundation for Business Execution – Jeanne W. Ross, Peter Weill, & David Robertson

• Enterprise Architecture Planning: Developing a Blueprint for Data, Applications, and Technology – John Zachmann

• “Enterprise Architecture, Organizational Structure” – Hamzeh AbuZaid

• Statistical Adjustment of Data – W. Edward Deming

• The Essential Drucker – Peter Drucker

• Patterns of Enterprise Application Architecture – Martin Fowler

• “Architect Your Business – Not Just IT!” – Jeanne W. Ross, Martin Mocker, Ina Sebastian

• “Enterprise Architecture: Driving Business Benefits from IT” – Jeanne W. Ross

www.linkedin.com/in/dustinhudsonkc