Agile Data Rationalization for Operational Intelligence

Welcome

Host:
Eric Kavanagh
eric.kavanagh@bloorgroup.com

Twitter Tag: #briefr The Briefing Room

Mission

!   Reveal the essential characteristics of enterprise software,
good and bad

!   Provide a forum for detailed analysis of today s innovative
technologies

!   Give vendors a chance to explain their product to savvy
analysts

!   Allow audience members to pose serious questions... and get
answers!


JANUARY: Big Data

February: Analytics

March: Open Source

April: Intelligence


Geoffrey Malafsky
Dr. Geoffrey Malafsky earned a Ph.D. in
Nanotechnology from Pennsylvania State
University. He was a research scientist at the
Naval Research Laboratory before becoming
a technology consultant in advanced system
capabilities for numerous Government
agencies and corporate clients. He has over
thirty years of experience and is an expert in
multiple fields including Nanotechnology,
Knowledge Discovery and Dissemination, and
Information Engineering. He founded and
operated the technology consulting company
TECHi2 prior to founding Phasic Systems Inc.,
where he is the CEO and CTO.


Agile Data Rationalization for
Operational Intelligence

Dr. Geoffrey Malafsky
Phasic Systems Inc

www.phasicsystemsinc.com
703-945-1378

2

Operational Intelligence and Data Rationalization
•  Operational Intelligence uses real-time data collected from
operating environments feeding analytical algorithms to detect
and predict problems and efficiency opportunities
•  It relies on and is vulnerable to:
▫  Data accuracy
▫  Data completeness
•  Big Data is really 2 types:
▫  Lots of data used for statistical analysis – quality is not critical
▫  Lots of data used for deterministic analysis – quality is critical
and high volume is limiting (CPU, storage, power)
•  Garbage in à garbage out; Big Garbage in à Galaxy Class
misinformation

3

Enabling Data Success
•  Overcome typical obstacles that prevented success in the past:
▫  Organizational group rivalry , Terminology confusion , Poor knowledge sharing ,
Inflexible designs
•  Rapidly build and manage data portfolio models that provides
visibility on strategy, stakeholders, designs, systems with
dependencies, linkages & analysis to operational data and metadata
•  Fill the gap in identifying, understanding and practically implementing
actual operational data versions with evolving standards and
consolidation
•  Distinguish, design, and implement similar, supposedly similar, and
operationally distinct data
•  Complement existing systems

Design Rationalization Issues System Rationalization Issues

•  Multiple data models •  Multiple database systems
•  Conflicting definitions •  Conflicting formats
•  Similar, supposedly similar, •  Redundant storage
operationally distinct values •  Unsynchronized values
•  Unknown business logic •  Multiple integration points
•  Multiple ETL mappings •  System performance

5

•  data values not metadata rule operations for application support, reporting, and decision making
•  data values are out-of-synch with all forms of metadata
•  data values conflict across data stores, organizational groups, and applications: syntactically
(simplest case) and semantically (most difficult)
•  top-down/bottom-up approaches have failed almost universally because they rely on metadata
and silo-ed organizational groups to solve what is inherently interrelated, complex
•  enterprise business goals are being hindered because of the poor data environment
•  there is little impetus to correct this situation
Different Meanings (Legal and
Business Activities)

NKY HomeSeekers Texas

6

Ψ-KORS Methodology: Data Rationalization and Portfolio Management
•  Integrated Organization,
Process, Technology
•  Synchronize metadata and
operational data
•  Allow valid, multiple distinct
versions of data entities
•  Cycle time in days/weeks
•  Correlated products

7

The Ψ–KORS™ System Model
Point-select data models, codes, entities

Data Rationalization
Design Rationalization System Rationalization
•  Consolidated, adaptive data models •  Consolidated, adaptive systems
•  Standardized definitions •  Common, interoperable formats
•  Synchronized distinct operational values •  Common storage
•  Managed business logic •  Synchronized interfaces
•  Coordinated ETL mappings •  Coordinated integration
•  Greater system performance

DataStar Discovery

DataStar Unifier

9

Corporate NoSQL™

Position Data Model

Perceptions & Questions

Analyst:
Eric Kavanagh


The Information Oriented Architecture (IOA)


Are We In the Data Tower of Babel?


Replace ‘God’ with ‘Innovation’ and…

God came down to see what they did and said: "They
are one people and have one language, and nothing
will be withheld from them which they purpose to
do." "Come, let us go down and confound their
speech." And so God scattered them upon the face
of the Earth, and confused their languages, so that
they would not be able to return to each other, and
they left off building the city, which was called
Babel "because God there confounded the language
of all the Earth".[3]


Modes of Transportation: I


Modes of Transportation: II


Modes of Transportation: III


Modes of Transportation: IV


The New Reality: I

!  Open-Source innovations are opening up whole
new ways of capturing, storing and processing
data; and many solutions are free, though you’ll
need trained developers to use the free stuff

!  Because the storage game has changed so much
with Hadoop, you can now store massive amounts
of granular detail, relatively cheaply

!  Big Data represents a huge opportunity, but also a
serious challenge for the business & IT


The New Reality: II

! NoSQL Database technologies change the game
due to greatly increased speed, among other
characteristics

!  Other innovations, including Massive Parallel
Processing, Multi-Core Processors and In-Memory
capabilities are also significant change agents

!  This opens the door to a new kind of information
architecture, with even real-time capabilities


The New Reality: III

!  The cost of software is in precipitous decline, as
evidenced by any number of metrics

!  In 2005, Microsoft quoted me $7,500 to host a
one-hour Webcast

!  In 2007, several vendors were offering pricing in
the $1,500-per-Webcast space

!  We now pay less than $500 per month for
unlimited Webcasts with WebEx


!  What is the NoSQL engine you’re using?
!  Could this replace both operational and analytical
Master Data Management solutions?

!  Is there any way to dynamically reconcile data
models? Or must you manually do this?

!  How do you deal with very old, “black box” legacy
systems?

!  Where would this sit in an information
architecture?

The Bloor Group

!  How do you deal with the User Adoption issue?
!  What would a small, foothold-style engagement
look like? What’s the low-hanging fruit?

!  You have a fascinating case study involving the
Navy and Human Resources Data. Can you describe?

!  Some consultants, like Michael Haisten in the 1990s
referred to an Enterprise Back Plane for data. That
was very similar to what’s now called Data
Virtualization. Do you see a comparison?

The Bloor Group

Mariah, tacked up and ready to sleigh! 
photo by pmarkham on Flickr 
 
Mangapps Railway Museum - 2009 
photo by Peter Taylor31 
 
xLamborghini Countach, Diablo SV and
Murciélago 
photo by exfordy on Flickr 
 
NASA SR-71B trainer after taking on fuel 
photo by jamesdale10 on Flickr

The Bloor Group

Thank You
for Your
Attention


Agile Data Rationalization for Operational Intelligence

Recommended

More Related Content

What's hot (20)

Similar to Agile Data Rationalization for Operational Intelligence (20)

More from Inside Analysis (20)

Recently uploaded (20)

Agile Data Rationalization for Operational Intelligence