TOGAF9 Quickstart Guide V10 C
TOGAF9 Quickstart Guide V10 C
TOGAF 9
Versiion 1..0c ((2009//11//18)) 3rd relleased Versiion Vers on 1 0c 2009 11 18 3rd re eased Vers on
ii
Disclaimer
While
the
author
used
his
best
efforts
in
preparing
this
book,
he
makes
no
representations
or
warranties
with
respect
to
the
accuracy
or
completeness
of
the
contents
of
this
book
and
specifically
dis- claims
any
implied
warranties
of
merchantability
or
fitness
for
a
particular
purpose.
The
advice
or
strategies
contained
herein
may
not
be
suitable
for
your
situation.
You
should
consult
with
a
professional
where
appropriate.
The
author
shall
not
be
liable
for
any
loss
of
profit
or
any
other
commercial
damages,
including
but
not
limited
to
spe- cial,
incidental,
consequential
or
other
damages.
Cover Picture
By
Wolfgang
Keller
2009
all
rights
reserved.
Location:
Berlin;
Hackescher
Markt;
Reflection
of
Berlin
TV
Tower.
Version History
Date
2009-11-18 Version 1.0c 2009-10-24 Version 1.0b 2009/07/27 Version 1.0a Includes Feedback from various colleagues. Special thanks to Gernot Starke. Have the permission to use original figures from TOGAF in the meantime but will stay with lookalikes for the moment as this results in a more consistent layout 2009/03/28 Version 09a First Version released. Any figures critical with respect to copyrights have been removed. Some might be replaced by TOGAF figures as soon as permission is granted. Feedback and corrections welcome. Dont ciriticize it help improve it! Language improvements by Andrew Black
iii
Preface
Why
would
anybody
need
a
60
pages
short
book
on
TOGAF
9
if
TOGAF
itself
is
a
780
pages
architecture
framework,
written
by
renowned
experts
from
300
top
IT
companies
who
sure
know
their
stuff
and
will
provide
you
with
anything
you
need
as
an
enterprise
architect?
The
answer
is:
YES
TOGAF
9
provides
you
with
very
helpful,
very
sound
and
extensive
lists
of
WHAT
to
do
in
Enterprise
Architecture.
The
advice
from
various
people
working
in
Enterprise
Architecture
is:
Use
TOGAF!
Theres
no
reason
not
to
use
it.
But
TOGAF
does
not
yet
provide
you
with
a
QuickStart
and
Accelerators
that
give
an
Expert
a
fast
to
read
ballpark
view
of
which
items
of
an
enterprise
architects
task
list
are
covered
by
TOGAF
and
which
are
not.
And
as
you
will
also
see,
there
are
areas
of
an
Enterprise
Architects
task
list
which
are
not
covered
by
TOGAF
at
the
moment.
This
makes
it
a
rewarding
task
to
give
people
interested
in
TOGAF
an
idea
of
what
they
can
expect
and
what
they
cannot
expect.
It
was
the
initial
plan
of
the
author
of
this
book
to
write
just
a
second
edition
of
his
own
book
on
IT
Enterprise
Architecture
which
has
a
substantial
share
on
the
German
market.
But
having
seen
the
new
TOGAF
9
document
and
taking
into
account
that
there
is
an
audience
of
about
10.000
certified
TOGAF
practitioners
out
there
and
maybe
far
more
than
100.000
IT
professionals
who
deal
with
TOGAF
world
wide
it
seemed
an
intersting
task
to
offer
a
Quick
Start
for
TOGAF
9
for
such
a
growing
audience,
interested
in
Enterprise
Architecture
and
interested
in
how
TOGAF
can
help
them
do
a
good
job
at
Architecting
the
Enterprise
iv
Table of Contents 1 Introduction 1.1 1.2 1.3 What is TOGAF 9? This Book is most useful for two Groups of Readers Your Feedback Welcome 1 1 2 3 4 4 7 9 9 10 11 12 13 14 16 18 18 19 21 21 22 23 24 24 26 26 27 30
2 What is Enterprise Architecture? 2.1 2.2 2.3 An Ballpark View of the Pragmatic Business Approach to Enterprise Architecture Work Breakdown Structure for Enterprise Architecture The Roots of the IT-Architecture Approach
2.3.1 Software Architecture as a Profession 2.3.2 Enterprise Architecture 2.3.3 Enterprise Architecture using the City Planning Metaphor 2.3.4 Preview: What has this got to do with TOGAF 2.4 2.5 2.6 A Few Notes on the Academic Approach The TOGAF 9 Approach Summary
3 TOGAF and IT Strategy 3.1 3.2 3.3 What is an IT Strategy? What needs to be described How do you get there?
3.3.1 Strategic Dialog 3.3.2 IT Maxims from Business Maxims 3.3.3 Reengineering the Business Strategy 3.4 3.5 What support will you find in TOGAF? Further Reading
4 TOGAF and IT Portfolio Management 4.1 4.2 4.3 4.4 What is IT Portfolio Management What is Application Portfolio Management? Putting it together: From Chaos to a Master Plan
What do you find about Infrastructure Portfolio Management in TOGAF? 33 Further Reading 34 36 37 38 38 39 40 41 42 42 43 44 46
5 TOGAF and Developing Architectures 5.1 5.2 5.3 Part II: TOGAF ADM Part III: ADM Guidelines and Techniques Further Reading
6 TOGAF and Architecture Governance 6.1 6.2 What do you find about it in TOGAF? Further Reading
7 TOGAF and Basic Tasks 7.1 7.2 7.3 TOGAF and finding the right Meta-Model for your Needs TOGAF and finding Tool Support TOGAF and Immersion Paths for Enterprise Architectures
Foundation Architecture / Technical Reference Model (TRM)46 The Integrated Information Infrastructure Reference Model (III-RM) 47 48 48 48 49 50 51
9 Wrap Up: TOGAF for You 9.1 9.2 9.3 9.4 A Collection of Useful Stuff The Two Strongest Points TOGAF Certification Future
10 References
1 Introduction
1 Introduction
1.1
What is TOGAF 9?
At the present time TOGAF, the Open Group Architecture Framework is a very popular framework for Enterprise Architecture (EA) world- wide. Version 9 the Version released in February 2009 is the result of a major rework of Version 8.1.1. TOGAF Version 8.1.1 was the latest update of the version 8 family which has been around since 2002, when the extensions for enterprise architecture were added to version 7 in the form of the so called enterprise continuum. Like many other frameworks in the IT architecture and manage- ment arena TOGAF 9 is not a use it right out of the box item. It has some 780 pages and is first of all a major challenge to read. Also, like many other IT frameworks, TOGAF will tell you more about WHAT to do than HOW to do it. Hence, when you start e.g. a new job as an Enterprise Architect and somebody passes you a copy of TOGAF 9 you might not really know what to do, unless you have had some other work experience besides TOGAF and have read a few other books. There is also the criticism that TOGAF does not provide you with working examples, document templates and other quickstart tools. Providing such material would indeed be very tough, as processes, documents and other material will differ depending on the kind of enterprise for which you have to develop an architecture. The good news is that TOGAF provides you with very useful checklists. The bad news is, that TOGAF will not provide you with an idiots guide to enterprise architecture. Even with TOGAF you are still allowed and required to think. Such an idiots guide does not yet exist and it is very unlikely that it will ever exist. Any approximation to it would consist of thousands of pages, would not be manageable and would be outdated by the time it was published.
1 Introduction
1.2
Experienced
EA
Professionals
who
are
not
yet
deep
into
TOGAF:
Will
get
a
quick
overview
on
where
to
find
what
in
TOGAF
9
and
where
TOGAF
9
has
still
gaps
if
compared
to
a
task
list
of
daily
enterprise
architecture
work.
People
from
this
group
should
be
able
to
read
the
book
in
a
cursory
style
in
something
like
an
hour.
Wherever
you
need
details,
you
can
dive
into
them
and
if
this
book
is
not
enough
you
will
get
a
pointer
into
the
respective
TOGAF
content.
To
make
a
long
story
short
for
such
readers:
TOGAF
is
sold
to
you
as
a
framework
for
Enterprise
Architecture.
It
is
very
strong
on
developing
architectures
but
it
is
still
somewhat
weak
on
Enterprise
Architecture
Management.
If
you
are
an
EAM
expert
and
you
know
e.g.
TOGAF
8.1.1
in
detail,
you
are
done
here.
TOGAF
9
is
not
much
better
at
EA
Managament
than
TOGAF
8.1.1
used
to
be
but
it
is
stronger
on
EA
Development
than
8.1.1
used
to
be.
You
can
go
on
flipping
through
the
book
in
order
to
draw
some
hints
from
the
short
recipes
and
further
reading
sections
that
you
might
not
have
come
across
before.
People
new
to
Enterprise
Architecture
and
interested
in
TOGAF:
Will
get
a
ballpark
view
of
what
to
expect
from
TOGAF
and
what
not
to
expect
from
TOGAF.
You
will
be
guided
through
a
process
framework
for
day
to
day
architectural
work
providing
you
with
an
overview
of
architectural
work
as
a
whole
and
you
will
get
information
on
which
of
these
processes
are
supported
by
TOGAF
9
and
the
level
of
quality
you
can
expect.
The
benefit
for
both
of
the
above
groups
of
readers
is
that
they
should
get
an
idea
of
how
they
can
profit
from
TOGAF
way
faster
than
by
reading
the
full
original
text.
But
they
will
still
not
get
the
worked
examples
and
all
the
supporting
material
that
would
pile
up
to
another
thousand
or
more
pages.
Also
because
such
material
is
of
limited
value
once
you
change
the
industry
or
move
on
to
an
enterprise
with
totally
different
IT
strategies.
If
you
are
an
experienced
TOGAF
expert
then
you
will
find
that
this
book
is
not
written
with
you
as
the
target
audience
in
mind.
You
are
better
off
with
one
of
the
migration
guides
for
transition
e.g.
from
Version
8.1.1
to
Version
9.
A
short
note
on
TOGAF
certification:
Having
worked
through
this
short
book
you
will
have
learned
that
TOGAF
certification
can
be
2009 Wolfgang W. Keller all rights reserved
1 Introduction
useful to prove that you did work through TOGAF and that you un- derstood the documents. TOGAF certification does not certify that you are a professional Enterprise Architect. Even though some HR departments may be made to believe this and even though there may be vendors for TOGAF courses out there who tell you this. Hence you can also use this book to find out quickly what TOGAF could give you and whether or not you think that it worth taking the exam.
1.3
The author acknowledges that this book might attract some critique from experienced TOGAF acolytes and also from people who have a marketing pitch that sells TOGAF as the one size fits all solution for Enterprise Architecture. Also some less experienced Enterprise Architecture professionals might be looking for the thousand page start-up material. Your feedback to improve this book is very welcome. The version you are currently reading has been written with the intention of sav- ing a lot of people a lot of reading time and also to confirm for them that they still have an appropriate notion of Enterprise Architecture Management even though they have the feeling that there is some- thing missing from TOGAF. This book will be distributed free of charge over the Internet while it is planned that the book will later be used as part of a regular book. Feel free to contact me with your suggestions for improvement.
As with building architecture there are many many defintions of Enterprise Architecture. In this chapter you will learn to recognise a few access paths to Enterprise Architecture and we will give a deeper explanation of a pragmatic access path that sees an Enterprise Architect as a very important aide to the CIO who is in charge, amongst other things, of strategic planning for the enterprises IT assets. We will call this the Pragmatic Business Approach. It will become clearer, once we outline the tasks of an Enterprise Architect as top aide to the CIO. The other two approaches are: The IT-Architecture Approach: This approach mainly has its roots in IT systems architecture. The systems architects often get to be in charge of a system cluster or even of a whole IT landscape and have tried to evolve their systems architects methods for use at the enter- prise level. You will learn that both the Zachman Framework and TOGAF have been greatly influenced by this approach. The Academic Approach: This approach first asks what is architec- ture of an enterprise and then looks for methods for modeling an enterprise, constructing meta-models for an arbitrary enterprise, evolution of meta-models and similar questions. Such questions focus on how to model an enterprise from top to bottom. In this chapter we will explain the three approaches and will position TOGAF 9 with respect to their viewpoints. In the section on the Pragmatic Business Approach you will become familiar with a list of top level tasks which will be used later to give you an idea of which tasks are supported by TOGAF 9 and which are NOT supported by TOGAF 9.
2.1
The Pragmatic Business Approach to Enterprise Architecture starts by asking what needs to be done to make the most of the enterprise resource IT. In more educated circles this is called IT / Business Alignment. IT / Business alignment is not a fully deterministic task. The way you try to pursue it depends on the maturity level of your organization.
Figure 1: Levels of IT / Business Alignment. There is a predictable set of tasks in a CIO office that serve to govern the expensive resource IT in an enterprise. The degree to which such a CIO office has implemented these depends on the importance of the production factor IT for the enterprise An enterprise which deals with commodity goods and does not see IT as some kind of factor relevant for its strategic positioning in competition will seldom ever have a complex CIO office. If it even has a CIO. A multi-billion business running an IT budget of several hundred million Euros a year will have all the functions that will be out- lined here maybe even more. In a typical CIO office you will find more or less the blocks of tasks outlined in Figure 2. Only two of these have something to do with enterprise architecture.
Figure
2:
Task
blocks
for
a
CIO
Office.
Dark
shaded
topics
are
not
sub- ject
of
EAM
and
will
hence
not
be
discussed
in
this
book
IT
Strategy:
Somebody
has
to
define
a
strategy
for
the
manage- ment
of
the
enterprises
resource
IT.
Project
Portfolio
Management:
Budgets
for
IT
have
been
around
much
longer
than
IT
project
architecture
or
even
Enter- prise
Architecture.
Thats
why
a
manager
like
a
CIO
used
to
have
somebody
in
charge
of
his
IT
budget
and
the
balancing
of
budget
and
demands
that
results
in
the
list
of
demands
which
are
to
be
implemented
each
year.
Enterprise
Architecture:
We
will
explain
the
tasks
an
Enterprise
Architect
has
to
perform
in
section
2.2.
IT
Audit:
As
the
IT
function
is
crucial
to
the
success
and
sheer
existence
of
most
enterprises
today,
IT
is
subject
to
audits
in
most
organizations.
Hence
there
is
a
task
to
audit
IT
functions.
Todays
de
facto
standard
for
auditing
IT
is
COBIT.
We
will
not
drill
into
this
any
deeper
as
it
will
not
help
you
too
much
in
understanding
TOGAF.
2009 Wolfgang W. Keller all rights reserved
IT Security: is crucial today for the reputation, integrity and security of your enterprise. Hence a CIO will have people in charge of IT security. Theres also a broad array of frameworks that will help you manage IT security. We will also skip this as- pect for the time being Provider Management: Most IT organizations today outsource or outtalk a lot of their work. This can apply to almost all tasks an IT organization performs, except the core management tasks. These core tasks are, by the way, described here for a CIO office. In the remainder of this book we will be dealing with the IT Strategy and the Enterprise Architecture blocks. Very often the Enterprise Architect, as a CIOs top aide, will help him define the enterprises IT Strategy. In any case, Enterprise Strategy and hence IT Strategy are the most important influences for you as an Enterprise Architect, even if your CIO does not consult you about formulating them. This and the elements of Enterprise Architectural work will be described in section 2.2. The structure explained here will be used later in chapters 3 thru 7 to explain what is in TOGAF to help you accomplish your tasks as an Enterprise Architect.
2.2
The tasks of Enterprise Architecture can be split into three main blocks as outlined in Figure Figure 3. Strategic Tasks: Often the Enterprise Architects are the people who help a CIO develop his IT Strategy. But besides this there are more strategic tasks meaning tasks that have a planning ho- rizon of more than 3-5 years. IT Portfolio Management will de- liver the basic data needed for Strategic Planning, which brings together the goals from strategy and the as-is situation from port- folio management in order to develop a to-be situation. This will be underpinned by a strategic roadmap that is a coarse program plan for a major part of the project portfolio.
Figure 3: Enterprise Architects Process Map Operational Tasks: form the day to day work of Enterprise architects. Strategies are nice but you need to communicate them and you also have to make sure that they are applied and implemented. This is the field of Architecture Governance. First of all you will need to find critical projects the ones that have the potential to change your architecture. This is done by Moni- toring the Project Portfolio. Once you have identified the 10% or so interesting strategic projects, you will set up an architects team to accompany them. As they run through the enterprises normal IT Project Process. Basic Tasks: In order to get Enterprise Architecture up and run- ning you will need to create a few foundations. In many cases it will be useful to run an EA tool in order to have a chance to track what you have in your application and infrastructure portfolios. In order to do this you will need to find or develop the right meta model for the purposes and in order to reduce complexity by standardization you will first have to develop standards that will be valid in the scope of your enterprise. There are more basic tasks but these are the most prominent ones. The following will explain each of the tasks listed above and in chap- ters 3 thru 7 you can find what material is available in TOGAF to help you perform them.
2.3
In this section we will give a short history of software and also of software architecture. The history of software development has pro- duced a set of abstractions and metaphors that over time gradually became stronger. This applies to all software related disciplines, including design of programming languages, platforms, processes, organizations and tools used to develop software. This is due to the fact that we are confronted with a quantity of software that is grow- ing exponentially since the term software itself appeared in the mid 1950s.
2.3.1
The
job
title
software
architect
is
very
en
vogue
today.
It
has
only
been
around
for
maybe
20
years.
Nowadays
it
is
tough
to
find
pro- grammers.
Most
people
in
the
trade
will
try
to
call
themselves
soft- ware
architects,
also
because
the
pay
is
much
better
and
because
the
title
is
more
appealing
than
just
programmer.
The
term
architecture
has
been
around
for
at
least
20
years
in
the
world
of
software.
Architecture
as
the
art
of
building
design
has
been
around
for
thousands
of
years.
There
was
architecture
even
before
people
built
the
pyramids.
Compared
to
that
software
itself
is
very
young.
At
the
beginning
of
the
1950s
software
was
not
even
an
inde- pendent
item
of
trade.
Software
as
we
know
it
today,
an
artifact
inde- pendent
of
the
hardware,
began
to
develop
in
the
ninteenfifties.
The
GUIDE
/
SHARE
organization
that
dealt
with
exchanging
software
for
IBM
computers,
was
founded
in
1955.
Following
that,
software
pro- jects
often
developed
operating
systems.
The
system
OS/360
was
a
typical
representative
of
such
projects.
Fred
Brooks
the
chief
de- signer
of
this
project
published
his
experiences
in
the
well-known
book
The
Mythical
Man
month
[Brooks75]
The
leading
figures
at
the
time
performed
Programming
in
the
Large1.
They
met
in
1968
at
the
famous
NATO
conference
in
Garmisch
in
order
to
think
about
how
to
cure
the
software
crisis
by
applying
the
new
metaphor
of
software
engineering.
Software
architects
were
not
to
be
seen
at
that
conference
or
at
that
time.
At
that
time
the
foundations
were
laid
for
techniques
such
as
data
abstraction
and
modularization.
The
research
was
mainly
focused
on
programming.
Instead
of
the
then
common
machine
languages
they
created
higher
level
abstractions
for
programming
in
order
to
deal
1
10
with the exponentially growing quantity of software. Programming languages that contained less powerful abstractions, such as COBOL appeared in the early nineteen-sixties, while more powerful lan- guages like ADA appeared around 1980. And still there were no soft- ware architects to be found in the programming crowd. A program- mer was still the thing to be. It was not until the mid nineteen-nineties that the term Software Architecture was documented in a systematic fashion [GarlanShaw96, Bass+98]. Software architecture like software engi- neering is a metaphor that tries to explain processes around the crea- tion of software using analogies to other disciplines of engineering in this case building architecture. People tried to transfer methods and processes from building architecture to the creation of software. At about the same time the Gang of Four [Gamma+95] created the foundation to describe recurring solutions in software using design patterns. It has to be noted that Gamma were not the first ones who transferred patterns from Christopher Alexanders [Alexander79a, Alexander79b] works to software design. You can find so called archi- tectural styles that form the repertoire of many software architects nowadays either as formal architectural styles as described by Garlan and Shaw [GarlanShaw96] or as architectural patterns as described by Buschmann et al. [Buschmann+96]. Software professionals, who were using these abstractions tended to call themselves software architects later on. From that point in time the profession of software architect, as we know it today started to evolve. This profession can be taught in seminars and university courses and there are also certi- fications for software architects. There is nowadays a reasonable measure of agreement on the curriculum for software architects. And as this group receives far better pay than just plain programmers you can observe a tendency of people to migrate into software archi- tecture instead of just plain programming.
2.3.2
Enterprise Architecture
The
growth
of
software
did
not
stop
at
the
border
of
single
(even
large)
software
systems.
The
metaphor
of
Software
Architecture
is
not
good
at
explaining
how
to
deal
with
complex
landscapes
that
are
composed
from
hundreds
or
even
thousands
of
major
software
sys- tems.
Hence
we
can
observe
the
evolution
of
a
new
metaphor,
called
the
City
Planning
Metaphor.
The
nineteen-eighties
and
nineties
saw
a
lot
of
so
called
stovepipe
architectures.
Systems
were
very
well
integrated
vertically
from
the
2009 Wolfgang W. Keller all rights reserved
11
user interface to the database layer. They were not so good (and often still are not so good) at horizontally integrating business processes. Horizontal integration often happened (and still happens) via the likes of data integration and batch data exchange. You might say: Long ago but remember that workflow systems have been around for roughly 15 years - in the perspective of building architecture this is just a blink of an eye. The people who pioneered these technologies in the nineteen-nineties did not even call them- selves software architects and no way enterprise architects. Nevertheless there were people practicing enterprise architec- ture at that time even though they did not know it. Terms like ap- plication portfolio that will be discussed in section 4.2 in this book did not exist in those days and are still not present in TOGAF (see section 4.4). If you are a fan of John Zachmans work you might state that he created the foundation for Enterprise Architecture in his famous paper [Zachman87, Sowa+92]. A closer look at that paper plus a look at the tasks enterprise architects perform today (see e.g. section 2.1) will show you that Zachmans work is project architecture for very large projects. If you search this work (and also TOGAF) for terms like Application Portfolio, Zoning Maps, and also Governance you get few to zero results. The City Planning Metaphor appears some 10-15 years later than John Zachmans work e.g. in a book by Longp [Longepe03] which is far less well known than the famous articles by Zachman.
2.3.3
If
a
Software
Architect
deals
with
a
single
system
or
maybe
a
cluster
of
7
or
10
systems
then
an
Enterprise
Architect
has
to
deal
with
the
set
of
all
applications
of
an
enterprise.
For
a
small
enterprise
this
set
may
have
50
applications,
for
a
major
insurance
company
the
figure
might
be
300
systems
and
a
global
car
manufacturer
typically
has
a
portfolio
of
more
than
3000
4000
IT
applications
individual
appli- cations
for
employees
workstations
not
counted.
An
Enterprise
Archi- tect
as
the
CIOs
city
planner
has
to
deal
with
almost
any
group
of
stakeholders
in
a
company,
be
it
senior
management,
be
it
groups
of
users,
or
be
it
public
agencies
like
the
SEC
(Securities
Exchange
Commission)
or
agencies
that
deal
with
data
protection,
and
last
but
not
least
those
groups
who
want
a
new
application.
As
these
tasks
have
some
similarities
to
city
planning,
the
profession
of
Enterprise
Architects
has
sometimes
adopted
that
term
for
their
occupation.
If
2009 Wolfgang W. Keller all rights reserved
12
you look at the zoning maps, city planners are using, then the analo- gies go even farther.
2.3.4
City Planners do not provide detailed plans for a single building. They will provide zoning maps that define which kinds of buildings will be arranged in which zone. They care for the set of all buildings of a city even though they would also have the qualification to plan a single one. In section 4.3 you will also get a glimpse of zoning maps for En- terprise Architecture and also other artifacts, which are frequently used by Enterprise Architects. In sections 4.1 thru 4.6 you will get a rather clear demonstration that TOGAF supports you with methods mostly derived from the Building Architecture Metaphor, but not with the later artifacts needed to help you when looking at your ap- plications as a portfolio of assets that needs to be managed. Given the fact that TOGAF has its roots in the DoDs TAFIM framework, which was handed over to the OpenGroup as early as 1995, this is some- what natural just looking at the period in which TOGAF and TAFIM were created.
13
2.4
As
one
sample
for
the
academic
approach
to
enterprise
architecture,
we
will
use
the
approach
pursued
by
the
(lets
call
it)
St.
Gallen
School
of
enterprise
architecture.
St.
Gallen
University
is
a
very
renowned
school
for
business
administration
and
also
business
oriented
com- puter
science.
This
school
uses
a
definition
of
Enterprise
Architec- ture,
which
is
also
promoted
by
Open
Group.
ENTERPRISE ARCHITECTURE: DEFINITION According to ANSI/IEEE Std 1471-2000, architecture is defined as the fundamental organization of a system, embodied in its components, their relationships to each other and the environment, and the principles governing its design and evolution (IEEE 2000). Enterprise architecture (EA) therefore is understood as (1) the fundamental organization of a government agency or a corporation, either as a whole, or together with partners, suppliers and / or customers (extended enterprise), or in part (e.g. a division, a department, etc.) as well as (2) the principles governing its design and evolution (OpenGroup 2003) . While an EA model is a representation of as-is or to-be architecture of an actual corporation or government agency, an EA framework provides (OpenGroup 2003) One or more meta-model(s) for EA description, One or more method(s) for EA design and evolution, A common vocabulary for EA, and maybe even Reference models that can be used as templates or blueprints for EA design and evolution. The components of an EA framework should be applicable for a broad range of corporations and government agencies. Source: Robert Winter, Ronny Fischer: Essential Layers, Artifacts, and Dependencies of Enterprise Architecture; [WinFis06]
If you see this definition you might think that the definition comes right from OpenGroup of TOGAF. This is not exactly the case. TOGAF and also the academic approach both like to cite IEEE 1471-2000 for the basics of Software Architecture. But the above definition is a compilation of various spots from TOGAF and way more rigorous than the sources it is drawn from. While TOGAF, as we will see in chapter 5 is interested in how you develop a concrete architecture, many academics are more interested in laying the foundations by e.g. dealing with questions of how to model an enterprise the right way. Some sample questions of current interest to the St. Gallen School include:
14
How do you model an enterprise as a whole from strategy via capabilities and business processes down to the level of infra- structure while being able to deduce properties or make state- ments about what is good or bad with respect to the strategic goals of an enterprise How do you find a proper meta-model for enterprise architecture in your enterprise How to you evolve or federate meta-models. From a research perspective all these questions are highly interesting and need to be asked with the long term perspective in mind that you want to do real Enterprise Architecture which means you do not only architect the IT part but the whole enterprise and you model Business and IT as a whole.
2.5
After
you
have
been
tortured
with
three
approaches
to
Enterprise
Architecture
and
since
this
is
a
book
on
TOGAF
it
is
somewhat
over- due
now
that
you
get
an
idea
of
how
TOGAF
sees
its
approach
to
enterprise
architecture.
Best
let
TOGAF
speak
for
itself:
TOGAF as an EAM Framework (From TOGAF 8.1. (Enterprise Edition) TOGAF in its Enterprise Edition remains what it has always been, namely an architecture framework - a set of methods and tools for developing a broad range of different IT architectures. It enables IT users to design, evaluate, and build the right architecture for their organization, and reduces the costs of planning, designing, and implementing architectures based on open systems solutions. The key to TOGAF remains a reliable, practical method - the TOGAF Architecture Development Method (ADM) - for defining business needs and developing an architecture that meets those needs, utilizing the elements of TOGAF and other architectural assets available to the organization. Source TOGAF 8.1. [TOGAF8.1.1] (Also valid for TOGAF 9)
So what does this tell us with regards to Enterprise Architecture tasks as described in section 2.2? The emphasis of TOGAF is on develop- ing concrete architectures: Be it for a system, be it for a cluster of systems and maybe also be it for a to-be architecture of an enterprise as a whole. Hence the emphasis is on the concrete task of developing architecture.
15
Figure
5:
Evolution
of
TOGAF
from
Version
7
to
Version
9.
Source:
Own
Research.
The
emphasis
of
TOGAF
is
not
on
tasks
such
as
Developing
an
IT
Strategy
based
on
a
Business
Strategy
Application
Portfolio
Management:
Dealing
with
thousands
of
existing
applications
and
judging
which
have
a
future,
which
need
change
and
which
need
to
be
replaced
Architecture
Governance:
This
is
mentioned
but
not
treated
as
a
main
item
The
ADM
(Architecture
Development
Method)
is
still
the
kernel
of
TOGAF
and
is
TOGAFs
biggest
asset.
You
can
also
see
this
concentra- tion
on
the
ADM
yourself
if
you
study
the
evolution
of
TOGAF
from
e.g.
Version
7
(published
in
Dec.
2001)
thru
various
versions
of
TO- GAF
8
(which
was
introduced
as
the
Enterprise
Version
of
TOGAF
7
and
first
featured
the
Enterprise
Continuum)
to
TOGAF
9
which
was
published
in
February
2009.
2009 Wolfgang W. Keller all rights reserved
16
Figure
5
can
be
read
so
that
Part
II
the
ADM
has
been
rather
stable
throughout
the
evolution
shown
in
Figure
5.
The
major
new
features
of
Version
9
over
Version
8.x
are
the
Architecture
Content
Frame- work
and
the
material
added
in
Part
III:
The
ADM
Guidelines
and
Techniques.
So
it
can
clearly
be
seen
from
this,
that
TOGAF
worked
its
way
up
from
a
framework
for
project
architecture
and
major
development
projects
to
the
enterprise
level.
TOGAF
will
continue
to
grow
towards
the
enterprise
level
but
it
has
not
been
designed
from
the
beginning
with
the
Enterprise
Level
or
a
Pragmatic
Business
Oriented
Approach
in
mind.
But
let
TOGAF
(again
8.1.)
speak
for
itself
again:
A number of enterprise architecture frameworks already exist and are widely recognized, each of which has its particular advantages and disadvantages and relevance - for enterprise architecture. They are discussed in Part IV (of TOGAF 8): Resource Base, Other Architectures and Frameworks. Although a number of enterprise frameworks exist, there is no accepted industry standard method for developing enterprise architecture. The goal of The Open Group with TOGAF is to work towards making the TOGAF ADM just such an industry standard method, which is neutral towards tools and technologies, and can be used for developing the products associated with any recognized enterprise framework - such as the Zachman Framework, Federal Enterprise Architecture Framework (FEAF), Treasury Enterprise Architecture Framework (TEAF), and C4ISR/DoD Framework - that the architect feels is appropriate for a particular architecture. . With the migration of TOGAF to an enterprise architecture framework, this flexibility becomes even more important. TOGAF is not intended to compete with these other frameworks; rather, it is intended to perform a unique role, in distilling what these other frameworks have to offer, and providing a generic ADM that can be adapted for use with any of these other frameworks. The Open Group's vision for TOGAF is as a vehicle and repository for practical, experience-based information on how to go about the process of enterprise architecture, providing a generic method with which specific sets of deliverables, specific reference models, and other relevant architectural assets, can be integrated. Source TOGAF 8.1. [TOGAF8.1.1]
17
2.6
Summary
So far we have seen that there are various approaches to Enterprise Architecture depending on ones background and objectives. The rest of this book will use the Enterprise Architects task list and will go into each major block of tasks, demonstrating what TOGAF has for you if you want to perform that kind of task and giving you pointers to additional material that you might want to use, when you are con- fronted with that kind of task.
18
In the early reactions to TOGAF 9 bloggers commenting on the new release have criticized TOGAF for falling short of providing help for the strategic planning tasks in IT management. The three main stra- tegic tasks have already been outlined in section 2.1 and IT strategy is the first one of them. A first check is quite simple: Just take the TOGAF pdf file, which is available at a modest charge from the OpenGroup and have the oc- currences of the term IT strategy counted in it. You will find the term three times in the text and this makes it clear that TOGAF does not provide you with a method, to develop one. If you just wanted to know, whether TOGAF provides you with a closed set of tools and methods to develop an IT strategy, youre done for the moment. Skip this chapter and GOTO chapter 4 on page 26 in order to see what TOGAF has to offer you for portfolio management. If you want to know where to find material on how to write an IT strategy and what TOGAF still has to offer besides obvious spots marked with IT strategy, hang on.
3.1
What is an IT Strategy?
In
order
to
talk
about
IT
strategy
it
is
first
necessary
to
talk
about
strategy
in
general.
In
order
to
avoid
copyright
problems
we
use
a
Wikipedia
definition
A strategy is a plan of action designed to achieve a particular goal.
You
could
also
use
the
more
militaristic
version
by
Clausewitz
[Clausewitz98]
but
theres
no
real
difference.
First
of
all
you
need
a
goal.
Then
you
need
more
than
one
possible
way
to
get
there
and
formulating
your
strategy
is
about
choosing
one
of
the
possible
ways
to
go
and
turning
it
into
a
plan.
You
might
ask,
whether
one
way
(or
option
for
attaining
the
ob- jective)
isnt
enough?
Many
strategies
do
not
pass
the
triviality
test
where
you
ask
whether
the
opposite
of
the
way
you
choose
is
still
2009 Wolfgang W. Keller all rights reserved
19
something that makes sense. E.g. if you do not hold a monopoly (you are in a competitive market) and you state Customer Orientation as your strategy, then the contrary being non-customer oriented would lead straight into bankruptcy. Hence for this example enter- prise Customer Orientation as a strategy would not pass the trivial- ity test because theres no alternative to it. You should find some indications of the strategy of your enterprise in its business strategy. We will also deal with the case where your enterprise doesnt have one. But lets postpone that one for a mo- ment. That case will be covered in section 3.3. You also need to know something about the maturity of your IT organization. If IT is treated as a pure non-strategic cost-center in your organization (remember Figure 1 for the different steps of IT / Business alignment of an IT organization) then you will have a tougher time influencing your enterprises strategy. When IT is seen as an enabler in competition, you should have less trouble entering into a true strategic dialog.
3.2
No matter at what level your IT organization finds itself the basics that need to be described (strategy elements) are always the same.
Figure 6: Main Pillars of an IT Strategy The above figure can use a bit of explanation. It is also reflected in the strategy matrix (see Figure 7) to follow. No matter what else you have in an IT strategy. You will always need to make decisions about the following:
20
Application Strategy: No matter what else you plan in your IT strategy, you need an application strategy a strategy that con- tains guidelines on how you want to deal with the IT applications in your company and how you want to support your business strategies using IT applications. Integration Strategy: In most cases you dont have one perfectly integrated application but you have a major set of applications that need to be integrated with each other in order to be able to perform business processes. These apps also might need to be connected to the outside world. SOA or EAI strategies are forms of integration strategies. But there are also more primitive ones. Infrastructure Strategy: You also cannot evade having an infra- structure strategy. Even if you outsource your complete infra- structure, you still have a strategy how to deal with it: in this case Outsourcing. Other factors that determine your infrastructure strategy is your global scope, your needs for security and special solutions and many other factors that you would check if you use a matrix like the one in Figure 7. Service Strategy: If you want to be perceived as a provider of services, you will end up having a service strategy, that deter- mines how your customers will get which services at which serv- ice levels plus some more decisions concerning Sourcing Strategy: Even if you should decide that you will pro- duce all your IT assets vertically integrated starting with develop- ing your own operating system: you still have a sourcing strategy. Normally you will make decisions here on what to outsource, and what to produce yourself. You will also decide whether you want to work together with a single provider, multiple providers, whether you want to source in an opportunistic fashion or relying on a few strategic partners to name a few of the decisions that need to be made here. If you define what you want to do and achieve in those 5 fields and if you can explain why this enables your business to pursue its Business Strategy, you are well off. In order to help you think about your deci- sions systematically and even deeper than that, you can apply key questions to each pillar the likes of how do I govern that specific area?; how do I finance it?, and a few more. As an example see Figure 7 below. This will force you to reflect on what you will do about certain general concerns, such as the regional distribution of your business.
21
Figure 7: Sample IT Strategy Matrix This matrix is like a checklist. If you are able to answer the questions in the matrix then you can be sure to a certain degree that you did not miss a major issue that you should have taken into consideration. You can then write down your five strategy elements. If youre good at it, you should have a short document of maybe less than 20 pages that informs your IT people about the strategic directions in each column and field.
3.3
The next question is: What is the process in which you need to in- volve Business that leads you to such a strategy document?
3.3.1
Strategic Dialog
The
best
way
would
be
if
your
CIO
and
a
few
key
IT
people
meet
with
key
Business
people
in
order
to
discuss
and
decide
on
a
Business
Strategy
that
also
includes
the
IT
Strategy.
This
state
of
affairs
is
somewhat
rare.
Analysts
state
that
about
90%
of
enterprises
do
not
have
a
written
business
strategy.
Often
CxOs
do
not
spent
the
time
for
a
proper
strategic
dialog
or
find
it
annoying
to
talk
to
IT
underlings
about
business
strategies.
In
such
cases
you
will
need
to
find
other
ways
to
extract
the
information
you
need
in
order
to
formulate
a
proper
IT
strategy,
which
will
lead
you
to
the
Maxim
Process.
2009 Wolfgang W. Keller all rights reserved
22
3.3.2
The Maxim Process is described by Broadbent and Kitzis in [Broadbent+05] as a pragmatic way to extract enough information for a good enough IT strategy while not investing more than a days workshop with senior management. The CIO will organize a work- shop with CxOs, which will lead to documenting 2 kinds of so-called Maxims: Business Maxims, And as a result IT Maxims Maxims are a few concise principles that are used to document the strategic direction of an enterprise. A Maxim workshop will usually not produce more than around 5 business maxims. For each of those, management will derive 4-5 maxims for the IT function that will help to support them.
Figure
8:
Maxim
Process
by
Broadbent
[Broadbent+05]
A
typical
Maxim
Workshop
will
be
split
up
into
two
parts:
Part
1:
Finding
the
Business
Maxims,
Part
2:
Deriving
the
IT
Maxims
An
external
facilitator
should
moderate
the
workshop
day
and
proc- ess.
To
give
examples
imagine
an
old
economy
financial
service
provider
like
a
big
insurance
company
that
runs
more
than
one
brand
name
on
the
market.
For
such
an
enterprise
you
could
find
the
following
busi- ness
maxim:
Create synergies in back office and service functions wherever brand identity is not compromised
IT
maxims
that
could
be
deducted
from
such
a
business
strategy
could
be:
(1) Define standard architectures and platforms used by all of the groups companies in order to leverage synergies and to reduce IT cost (2) Harmonize the IT application systems for the groups companies wherever there is a business case for this. 2009 Wolfgang W. Keller all rights reserved
23
Such a synergistic strategy (or any other strategy) will have implica- tions on ones IT governance principles. Your set of IT maxims defines the IT objectives and the chosen ways for reaching them. (See also Figure 8) You will need to install a system of IT governance that matches your IT maxims. You might say: Easy and straightforward process. There should be no problem to hold a meeting, define a few IT maxims and off you go with a usable IT strategy. Life is not always that easy. There are more than a few things that can go wrong here: One we mentioned already above: If the CxOs see IT as a bunch of underlings and not as an important provider of business oppor- tunities, they will not even invest the time in such a workshop. Another potential threat is less evident. If the board of CxOs has a culture of hiding conflicts and if they do not agree on a common strategy they will not be too interested that their internal brawl becomes known throughout the enterprise. In such a case you will never get the maxim workshop for whatever reasons. This will lead you to the next level. You will need to reengineer an IT strategy from the behaviors you observe in daily business. Which will bring us to the next section.
3.3.3
If
your
management
does
not
want
to
spend
the
time
to
discuss
Busi- ness
Strategy
and
IT
strategy
with
you,
its
time
for
you
to
reengineer
the
strategy
from
what
you
experience.
There
are
enterprises
around
who
are
very
successful
and
have
a
strategy
or
call
it
a
very
successful
way
of
operating:
But
its
neither
written
down
nor
could
anybody
explain
it
explicitly.
In
such
a
case
you
need
to
reengineer
or
find
the
business
max- ims
by
applying
analogies
with
known
strategy
patterns.
Literature
will
provide
you
with
an
ample
body
of
knowledge
on
Business
Strat- egy
and
will
enable
you
to
identify
the
patterns
comprising
your
enterprises
strategy.
Here
again
a
matrix
like
the
one
presented
in
Figure
7
is
useful.
You
can
ask
questions
like
what
is
the
geographic
distribution
of
my
enterprise
and
what
are
best
practice
patterns
to
deal
with
just
that
kind
of
geographic
distribution
when
it
comes
to
services,
applica- tions,
infrastructures
or
sourcing.
That
way
you
can
again
go
through
the
matrix
(Figure
7).
2009 Wolfgang W. Keller all rights reserved
24
3.4
As you can see from the TOGAF overview picture2, TOGAF treats Business Strategy and also IT Strategy as an item outside of the scope of TOGAF. Phase A (Architecture Vision) of the ADM [TOGAF9, Chapter 7, pages 81ff] tells you that you should get hold of the business drivers before you start an architecture project of some kind. TOGAF also proposes a capability analysis. Still the combination of Business Strategy and IT Strategy seems to be considered as something outside the scope of at least the TOGAF framework.
3.5
Further Reading
Gartner Material on IT Strategies Gartner usually has some interesting material on IT strategies. We cannot cite the stuff here because their Intellectual Property policy prohibits citations of material older than one year. Have a look at the Gartner Research databases if you happen to have a relation with them. Marianne Broadbent used to be a Senior Vice President at Gartner, and also served as a professor at Melbourne Business School. She has consulted and interviewed a four-digit number of CIOs for various panels by Gartner. Her book New CIO Leader: Setting the Agenda and Delivering Results [Broadbent+05] is a classic on IT strategy; MIT Sloan School Material on IT Strategies Gartner is also connected to the MITs Center for Information Systems Research (CISR) headed by Peter Weill. A very interesting book is Enterprise Architecture as Strategy: Creating a Foundation for Busi- ness Execution [Ross+06] by Jeanne Ross and Peter Weill. The book will also point you to further MIT resources or check the MIT website at http://mitsloan.mit.edu/cisr/.
For the respective TOGAF picture see http://www.opengroup.org/architecture/togaf9-doc/arch/Figures/01_structure.png The author has to ask for permission before reprinting the material.
25
Consulting Tools If you develop strategies it very helpful to have a working knowledge of methods and tools that were developed by the likes of BCG, McKin- sey and other well renowned consulting firms. A convenient way to get such information is to get yourself books on consulting toolkits, strategy consulting or just a compact MBA book for starters. Theres too much such literature around to be counted.
26
TOGAF has almost nothing to offer for IT portfolio management. Hence we will give a short summary of Application and Infrastructure Portfolio Management and will set pointers to the few spots in TO- GAF, which are of some help for these tasks.
4.1
IT
Portfolio
Management
comes
in
at
least
three
flavors.
Two
of
them
are
relevant
for
Enterprise
Architects.
And
the
third
flavor
is
also
present
in
a
CIO
office
and
hence
Enterprise
Architects
will
at
least
have
an
interface
to
it.
The
flavors
are:
Project
Portfolio
Management:
Of
all
the
flavors
of
portfolio
management,
this
is
the
most
widespread
one.
Even
if
companies
dont
care
about
architecture
they
usually
care
about,
how
their
money
is
spent.
It
is
quite
a
common
situation
that
a
company
has
two
or
three
times
more
budget
proposals
for
projects
than
it
has
money
to
spend
on
projects.
This
explains
why
a
mechanism
for
project
prioritization
is
installed
in
almost
any
enterprise.
As
stated
above,
Project
Portfolio
Management
has
an
interface
to
Enterprise
Architecture
but
is
usually
not
seen
a
part
of
it.
This
is
why
this
subject
is
not
treated
any
deeper
in
this
book.
Application
Portfolio
Management:
Most
Enterprises
have
a
considerable
number
of
applications
that
support
their
business
processes.
Below
you
will
learn
why
it
is
important
to
have
an
in- ventory
of
applications
and
what
goals
managing
them
pursue.
Application
Portfolio
Management
is
usually
seen
as
a
part
of
strategic
Enterprise
Architecture
Management.
Therefore
it
will
be
treated
below
in
some
more
detail.
Infrastructure
Portfolio
Management:
Often
Enterprises
do
not
have
a
proper
IT
inventory.
Nevertheless
some
of
those
who
do
not
have
a
99%
accurate
IT
inventory
still
manage
their
Infra-
2009 Wolfgang W. Keller all rights reserved
27
structure Portfolio. The reason behind this is in most cases cost savings by standardization. Which can be translated to: The less technologies you own, the lower your maintenance and admini- stration costs. For most companies IT infrastructure is not a mat- ter of competitive advantage. Therefore they can afford to man- age it as a commodity where the cheaper solutions (fewer tech- nologies, fewer locations) are often the better solutions. The rest of this chapter will explain Application Portfolio and Infra- structure Portfolio Management. We will start with Application Port- folios, just because this is handier to explain. Infrastructure Portfolio can often be even more rewarding as Enterprises usually spend much more money on IT infrastructures than they spend on applications.
4.2
In
order
to
understand
why
you
would
like
to
deal
with
portfolio
management,
it
is
somewhat
helpful
to
explain
the
terms
Lets
start
with
the
term
portfolio.
A
portfolio
is
a
collection
of
in- vestments
held
by
an
institution
or
a
private
individual.
This
means,
in
a
normal
enterprise
you
have
an
IT
infrastruc- ture
portfolio
and
also
an
IT
application
portfolio,
because
you
simply
possess
a
set
of
infrastructure
components
and
also
a
set
of
IT
appli- cations.
And
usually
you
have
heavily
invested
in
them.
Next
lets
have
a
look
at
portfolio
management
in
finance.
A
port- folio
manager
will
(in
theory)
optimize
his
portfolio
so
that
it
yields
a
maximum
return
at
a
given
level
of
risk.
Usually:
The
higher
the
risk
the
higher
the
potential
gain.
And
this
is
about
the
end
of
the
analogy
for
several
reasons:
Measuring
Returns:
For
financial
assets
it
is
comparatively
easy
to
measure
a
monetary
return.
You
get
dividends,
interest
or
other
forms
of
payments
and
also
the
value
of
your
asset
may
change.
For
an
IT
application
it
is
tough
to
measure
the
return,
because
it
is
tough
to
determine
which
part
of
your
companys
success
can
be
attributed
to
the
IT
application
and
which
to
other
factors.
No
Exchange:
Usually
there
is
no
stock
exchange
for
IT
applica- tions.
Investments
in
a
financial
portfolio
can
usually
be
traded
on
a
stock
exchange
or
some
other
exchange.
There
are
brokers
who
deal
in
used
IT
infrastructure.
But
for
individual
used
appli- cations
there
is
nothing
similar
to
a
stock
exchange
or
other
mar- kets.
Individual
IT
applications,
which
make
up
the
lions
share
of
2009 Wolfgang W. Keller all rights reserved
28
your IT investment in application systems, tend to have a low (if not zero) fungibility. Covariance: Have you ever tried to sell the Austrian IT applica- tion for the cadastral register to lets say the Chinese govern- ment? You will experience problems that stem from use of other applications that your cadastral register application is interfaced with or is based on. For example, maybe, the local authority regis- ters of residents. Often an enterprise application needs a whole ecosystem of other applications it works with. In an investment portfolio you can sell XY stock and replace by YZ stock. In an ap- plication portfolio such a switch usually results in expensive and long-term projects. Then why would people still call managing their applications Appli- cation Portfolio Management if the analogies are limited? The an- swer is because a lot of the techniques used to analyze application portfolios use 4-quadrant matrices in the style of the famous BCG matrix.
Figure
9:
BCG-style
Matrix
used
to
analyze
Application
Portfolios
Similar
matrices
can
be
found
in
Ward
/
Peppard
[Ward+02]
2009 Wolfgang W. Keller all rights reserved
29
The
categories
given
by
Figure
9
are
just
one
of
many
many
ways
to
look
at
your
collection
of
applications
though
it
is
a
particularly
common
and
popular
one.
The
applications
you
find
in
each
quadrant
require
a
distinct
treatment
that
is
common
for
all
the
applications
in
a
quadrant.
Dogs
(a.k.a.
Support
Applications):
These
are
those
applications
that
have
a
comparably
low
contribution
to
todays
business
results
and
also
a
low
contribution
to
future
business
results.
E.g.
a
general
ledger
system
is
not
a
critical
success
factor
for
most
companies
and
it
will
never
be
one.
Hence
it
is
a
common
strategy
to
identify
such
systems
and
make
an
analysis
whether
the
company
gets
a
better
deal
if
you
outsource
such
systems
completely.
Not
only
development
but
also
operations.
Cows
(a.k.a.
Key
Operational
Applications):
These
are
the
proc- esses
your
current
bottom
line
relies
upon.
In
many
companies
these
would
be
large
individual
software
systems
with
a
heavy
investment
bound
in
them
and
very
conservative
procedures
to
change
them.
Downtime
in
such
systems
usually
has
a
direct
effect
on
the
bottom
line.
The
company
usually
could
not
afford
that
and
hence
has
to
concentrate
a
lot
of
attention
on
keeping
these
systems
up
and
run- ning.
Outsourcing
the
whole
system
is
seldom
an
option.
Question
Marks
(a.k.a.
High
Potential
Applications):
Typically
you
have
a
research
department
(or
a
similar
organization)
that
takes
care
of
you
product
innovations.
If
you
invent
new
products,
the
sys- tems
that
come
along
with
them
need
not
be
extra
stable
but
only
need
to
be
demonstrators
that
allow
your
management
or
customers
to
assess
the
possibilities
in
a
product
or
system.
Usually
you
would
have
many
such
test
systems,
as
maybe
1
out
of
10
innovation
really
makes
it
to
the
market.
Stars
(a.k.a.
Strategic
Application):
These
are
the
applications
that
belong
to
products
that
have
been
selected
to
be
the
future
cash
cows.
Such
products
should
be
in
a
heavy
growth
phase
with
small
absolute
numbers
sold
but
high
growth
rates,
often
in
very
competitive
and
also
growing
markets.
The
emphasis
here
is
typically
on
speed
and
features
and
less
on
an
absolutely
minimal
number
of
production
problems.
In
many
cases
you
would
want
to
keep
the
systems
that
support
such
products
apart
from
the
systems
that
support
the
cash
cows.
But
there
are
also
industries
like
Telecoms
where
you
have
to
test-drive
your
innovations
with
the
big
iron
systems,
as
there
is
only
one
system
that
allows
you
to
do
billing
for
all
kinds
of
products.
Using
such
analyses
leads
to
some
interesting
other
views
on
a
port- folio
e.g.
which
quadrants
does
your
project
money
go
to?
The
anti- patterns
are
so
instructive
that
most
analysis
here
is
straightforward.
2009 Wolfgang W. Keller all rights reserved
30
E.g. if you spend 60% of your whole project budget in the question marks quadrant on three huge projects, experience should tell you that something is wrong, as spending in the Question Mark quadrant should be relatively moderate. Or if you have a major number of pro- jects in the Dogs quadrant then you should ask yourself, whether it makes sense to nurture the poor dogs by additional projects instead of outsourcing them.
4.3
If you get assigned to the new job of Chief Enterprise Architect, you will often find no inventory and no analysis. The question in such a situation is: How do I construct a proper To-Be state of my applica- tion landscape and how do I produce the master plan that gets me from my As-Is state to a proper To-Be state. Figure 10 gives you an overview of the analysis elements that are often applied.
Figure
10:
Elements
of
an
Application
Portfolio
Analysis
Application
Handbook:
In
most
cases
you
start
by
creating
an
inventory
of
all
your
applications.
Often
you
will
also
draw
proc- ess
support
maps
that
show
you
how
your
business
processes
are
supported
by
applications.
This
will
give
you
an
overview
of
what
you
have.
But
it
does
not
yet
really
help
you
to
assess
what
is
good
and
what
is
bad
with
respect
to
your
business
goals.
Heat
Mapping:
A
further
step
is
often
a
so-called
Heat
Mapping.
This
is
also
available
as
a
service
offering
called
Motion
from
Microsoft
for
a
majority
of
industries.
An
expert
will
bring
a
list
of
potential
capabilities
for
your
industry
and,
together
with
a
team
of
experts,
you
mirror
potential
capabilities
against
your
business
strategies
and
you
will
find
out
which
are
the
critical
capabilities
that
you
need
to
specially
look
after
in
order
to
implement
your
2009 Wolfgang W. Keller all rights reserved
31
strategy. You can then assess the actual quality of the capabilities as provided by your application portfolio and will get a per capa- bility heat map that is suited to demonstrate the areas for poten- tial action quite intuitively. Reference Models: Often it is very instructive to map your appli- cation landscape against domain specific reference models. Often you will find a bunch of areas for improvement Portfolio Analysis: And you can also use Portfolio Analysis to cluster your applications into classes that usually deserve some kind of standard treatment like outsource, stabilize, reno- vate and other treatments. If you combine the results from those four streams of analysis it is in most cases straightforward to come up with ideas for action, like You find that for a critical capability you have 5 different systems in 9 locations that support it. In such a case you might want to consolidate the apps and replace them with the one that best or better supports your critical capability. You find that a relatively huge system supports 8 capabilities, 3 of them critical and 5 of them non-critical commodities. You might want to come up with ideas on how to separate the commodity part from the critical part and construct better-suited system support at lower costs. The advice here is rather generic. The reason for this is that you can describe a generic method like heat mapping or use of process sup- port maps but it is difficult to foresee the outcome for each possible enterprise and describe a deterministic cookbook that shows how to attack any problem the world of application planning has to offer. Your To-Be architecture will be an array of transformations applied to your As-Is architecture. The choice and quantity is determined by your total budget and also priorities. The selection process can be copied from ordinary project portfolio management. Figure 11 shows representatives for the two additional artifacts that your master plan will contain: You will have To-Be application maps (e.g. again process sup- port maps) of your target application landscape And you will have a master plan showing the operations of the application landscape e.g. as project plans or interval maps that show which system is to be replaced by which other system and when.
32
You could also add the planned state of your heat maps your to-be state or the planned state of your portfolios.
4.4
The answer is that Application Portfolio Management, as a term does not appear at all in TOGAF. As mentioned above, TOGAF is not really intended to help you with strategic IT management and therefore TOGAF is not the prime source to use if you need to learn something about strategic IT management.
4.5
Just
as
you
can
manage
applications
you
can
also
manage
your
infra- structure
components,
most
likely
consisting
of:
Your
network
infrastructure
consisting
of
WANs,
LANs
and
other
network
components;
Servers
(from
PC
server
clusters
to
mainframe
computers);
And
all
kinds
of
infrastructure
services,
like
operating
system
services
(be
they
virtual
or
real)
and
many
other
basic
services
like
archiving,
or
printing
There
are
at
least
two
differences
here
to
application
portfolio
man- agement:
You
will
most
likely
manage
classes
of
devices
and
services
in- stead
of
single
instances
(applications).
It
is
not
the
target
of
in- frastructure
portfolio
management
to
replace
a
CMDB
(configura-
2009 Wolfgang W. Keller all rights reserved
33
tion management database). You may therefore work on a clus- tered extract of a CMDB and your portfolio will most likely con- sist of a number of services and heterogeneous implementations thereof. In short, you will find too many implementations for the same service, too many operating systems and variants, too many software products for the same task and your job will most likely be to save money by reducing heterogeneity. As most infrastructure components in most enterprises are commodities your main interest in them will be cost reduction while guaranteeing a given level of service. Hence you will in most cases skip a lot of the dimensions that you would deal with in Application Portfolio Management and go primarily for dimen- sions such as cost, homogeneity and simplicity of a portfolio. So the most likely form of report or architectural artifact in Infra- structure Portfolio Management is a simple list, which lets you see how many implementations of service X you provide and which al- lows you to work out how to reduce the complexity of your portfolio.
4.6
The straight answer is: If it comes to the management aspect you will find almost nothing. What you can use is TOGAF terminology on in- frastructure, as described in the Technical Reference Model. This will give you terms and taxonomy of classes of infrastructure and also infrastructure services. For the top level of this see e.g. Figure 12. If you say trivial then please do not forget the last time consuming discussions you had when trying to agree on such a picture and the terms behind it. So the value of such a de-facto norm should not be underestimated.
34
Figure
12:
Top
Level
View
of
the
TOGAF
Technical
Reference
Model.
There
are
deeper
levels
to
this
and
also
taxonomies.
Relevant
Layers
for
Infrastructure
Portfolio
Management
are
Applications
Platforms
and
also
Communications
Infrastructure.
For
the
similar
figure
in
TOGAF
9
see
figure
43-1.
http://www.opengroup.org/architecture/togaf9-doc/arch/Figures/43_trm.png
You can also use the TOGAF Content Metamodel3 in order to model your infrastructure. You will find metamodel snippets for Platform Services, Logical Technology Components, and also Physical Technology Components, Which will help you with a first cut of a set of entities and attributes you can use for your own infrastructure portfolio management.
4.7
Further Reading
It
is
somehow
tough
to
find
a
hands-on
guide
to
Application
and
In- frastructure
Portfolio
Management.
What
you
will
get
is
fragments
but
you
will
have
to
integrate
them
yourself
at
the
moment,
starting
from
the
questions
you
and
your
management
want
answers
for
and
then
designing
the
analysis
instruments
to
answer
them.
The
classic
for
Application
Portfolio
Management
is
the
book
by
Ward
and
Peppard
Strategic
Planning
for
Information
Systems
[Ward+02]
which
has
appeared
in
different
editions
even
with
changes
in
the
authors
team
over
the
years.
The
book
is
not
easy
reading
but
a
classic
and
worth
owning
if
you
have
concrete
tasks
around
Application
Portfolio
Management.
Kaplans
book
on
Strategic
IT
Portfolio
Management
[Kaplan05]
has
some
kind
of
focus
on
Project
Portfolio
Management,
but
is
one
of
3
35
the few books on the market today that deals with IT Portfolio Man- agement at the title level. TUM EAM Pattern Catalog: A group at the Technical University of Munich (Germany) collects so-called EAM patterns. Here you find management procedures, viewpoints used for them and also meta- model snippets that support them. You could also start from the questions, look for right patterns and then integrate your personal version of a portfolio management from them. Just have a look at http://wwwmatthes.in.tum.de/wikis/sebis/eampc. Looking at the site yourself is much faster than reading a third party text like this one. German Books: If you read German you will an array of books that emphasize the management aspect of Enterprise Architecture Man- agement like [Dern09, Keller06]. If you are especially interested in application portfolio management, it is worth having a look at the MS thesis work by Riege [Riege05]. A relatively new book by Inge Han- schke is especially focused on applying Zoning Maps to strategic planning of application landscapes [Hanschke08].
36
From time to time even Enterprise Architects need to develop archi- tectures. Those can have such different scopes as An architecture for a single system that consumes an effort of maybe only a few person years Clusters of systems that support a single business process, span- ning maybe 5 7 IT systems. Such projects may consume 10 to 50 person years. Architecture for the application landscape of a whole company. This will never be implemented in a single project a may take al- most a decade to complete. Template Architectures and architectures for application frame- works that serve as the template for a family of applications, like e.g. all customer facing web applications of an enterprise. Such BluePrints are typically reused 5 10 times. Such tasks are the core competence of TOGAF and this is also why this chapter will be short because we will forward you to TOGAF after a short look at what TOGAF offers you. The TOGAF ADM (Archi- tecture Development Method). If you know a TOGAF version the likes of 7.0 or 8.x youre done here with this chapter. Have a look at how TOGAF has improved the ADM by factoring out a few things and splitting it into two parts: Part II and Part III. If you happen to be a TOGAF newbie hang on for 2 pages.
37
5.1
TOGAF proposes a cyclic process for architecture development (see Figure 13). What you have to do here is quite straightforward for any software architecture professional.
Figure
13:
Cyclic
ADM
Process.
For
the
similar
figure
in
TOGAF
9
see
figure
5-1.
http://www.opengroup.org/architecture/togaf9-doc/arch/Figures/adm.png
You can see the steps as offerings or as a checklist. If you would e.g. make a minor adjustment to a few systems within a given Business Architecture you what have a look at the check list for Phase B in the cycle and would skip it, if you come to the conclusion that what you would have to do here is already clear or has been done by other colleagues. Each of the Phases at the top level is depicted in Figure 13 Each phase is split up in various steps that make a veritable checklist for the phase. TOGAF provides information on documents that should go into a step and should be produced in each step.
38
5.2
With TOGAF Version 9 a lot of complementing material has been factored out of the core ADM cycle. Therefore Part III contains all the additional material that is there to help you using the ADM such as information on: How to use the ADM as a cyclic process (chapter 19) Applying the ADM at different levels of granularity (smaller or larger projects) (chapter 20) And such aspects as doing security engineering, using TOGAF for a SOA (chapter 22), stakeholder management (chapter 24), archi- tecture patterns (chapter 25), and a few more useful things for architects to know. Part III is therefore more like a collection of useful stuff than a check- list or a process like Part II the core ADM
5.3
Further Reading
Even though TOGAF is a very extensive and useful checklist for soft- ware architects it is more of a checklist than training material. For example, if you want to apply requirements management (the central item in the middle of the cycle) you need to have knowledge of re- quirements management and it should be more knowledge than can be found in TOGAF as the more extensive books on Requirements Managements have their own 500+ pages compared to the 780 pages TOGAF has altogether. Or take chapter 25 on architecture patterns. You will get a few references to the likes of the IBM eBusiness Patterns or other pattern sources and will get an idea of what a pattern is and what it is useful for. But you cannot expect a full catalog of all patterns an architect could or should - know. Luckily TOGAF has some 3 pages of references and links to litera- ture, so that it does not make sense to cite another page of references for each of the chapters 5 thru 32. Still a few architecture books are quite helpful as the references contain more pointers to other stan- dards and frameworks than to good introductory text. Which can be translated as: The ADM is not an introductory text to software architecture for beginners but TOGAF is a condensed checklist for architecture professionals.
39
If you remember Figure 3: The Enterprise Architects Process Map then you will remember that an Enterprise Architecture Group has three big groups of tasks: The strategic Tasks have been discussed above in chapters 3 and 4. The architecture development tasks have been be discussed in chapter 5. What remains is the question of how to enforce the strate- gies and architectures that you have developed. This is called Archi- tecture Governance. For the further discussion we will use a part of Figure 3
Figure 14: Architecture Governance Overview So in order to exercise practical Architecture Governance you need to do the following: First of all you need an actual overview of the project portfolio of your enterprise. You are not interested in those projects that per- form simple maintenance tasks or are the fifth implementation of a standard blueprint. You are interested in those projects that do things that change your overall architecture and that have an im- pact on your future IT landscape. So your first task is, to find ex- actly those interesting projects.
40
The next important point is to get an idea of whether the team that is assigned to the project that you found interesting, is capa- ble of performing a proper job. If you find that an inexperienced architect was assigned to very difficult job, the project will need far more attention and investment from your side, than it would have needed if your counterpart were a highly professional and well-educated architect. After screening the project, and after getting an idea of what you have to expect from the team, you would agree on an audit plan for the project. Depending on the degree of difficulty, you can set up the frequency of meetings necessary to govern the project. In those meetings you will discuss progress with the projects archi- tects and you will also discuss deviations from the company ar- chitecture policy. As you may see, all this has a lot to do with communication and the ability to judge your colleagues capabilities to come up with the right solutions the first time.
6.1
In TOGAF you will find all the technical definitions for IT governance and also for architecture governance. You will not find the tips and tricks you need to deal with your colleagues, the techniques for effec- tive reviews, or what you have to do in cases of deviations. TOGAF has two spots that deal with architecture governance: First, there is a dedicated chapter, chapter 50, on architecture governance. This chapter contains mostly the scope and defini- tions needed to perform architecture governance beyond what was outlined above. TOGAF also sees enforcing compliance as a task for architecture governance activities. Second, phase G of the TOGAF ADM (chapter 15 of TOGAF) covers Implementation Governance. Here you find a more or less de- tailed recipe on steps needed to perform an architecture audit. So in addition to a few tricks on how to handle reviews you also find sufficient information.
41
6.2
Further Reading
Besides the more formal advice on architecture governance, audits, and reviews, there is a lot of useful advice on how to improve practi- cal architecture work in the pattern community. You can, for example, use the so-called writers workshop in order to improve almost any kind of documentation on architecture work. For a guide on how to conduct the writers workshop consult work by Jim Coplien [Coplien96, Coplien97].
42
TOGAF offers some help, if it comes to other architecture routine tasks. As an architect you will sooner or later want to store the infor- mation you acquired in some form of automatic database, instead of in a bunch of Excel sheets and PowerPoint presentations. For this you might need either a software tool or a meta-model. In most cases you would need both. Sections 7.1 and 7.2 will deal with how to de- sign a meta-model for use in enterprise architecture and also what you can expect from TOGAF if it comes to selecting the right tools for Enterprise Architecture Management. Apart from that we will give you a quick overview in section 7.3 of what you can expect from TOGAF if you want to set up an enterprise architecture function.
7.1
In
Enterprise
Architecture
Management
you
sooner
or
later
come
to
a
point
where
you
want
to
have
a
database
of
your
IT
Portfolio,
your
IT
assets
and
your
application
landscape
(to
name
a
few
items)
in
order
to
perform
management
on
the
sets
of
real
world
items
you
have
modeled
in
your
Enterprise
Architecture
Database.
You
can
get
ready
to
use
meta-models
in
sizes
between
50
meta- entities
up
to
far
more
than
500
meta-entities.
It
should
be
somewhat
straightforward
to
see
that
50
is
somewhat
limited
and
more
than
500
might
be
a
bit
too
complex
if
not
covered
by
a
very
well
inte- grated
user
interface
of
an
EAM
Tool.
In
many
cases
you
will
not
try
to
answer
all
possible
questions
in
EAM
at
a
time.
It
is
more
likely
that
you
management
has
focused
its
interests
on
certain
points
like
e.g.
cost
management
for
your
infra- structure,
just
to
name
an
arbitrary
example.
Start
small:
In
such
cases
it
is
important
to
start
with
a
small
so- lution
that
is
driven
by
the
question
you
have
to
answer
without
having
to
deal
with
the
full
complexity
of
a
500+
item
meta-model.
2009 Wolfgang W. Keller all rights reserved
43
You can still end up big: TOGAF architecture capability frame- work offers you a meta-model, which is split or modularized into small areas of interest. If you are interested in a certain area, you need to read the respective sections and can draw from the lists of predefined TOGAF meta-objects. The top-level picture of the TOGAF Architecture Content Framework gives you an overview of the areas for which you can expect support.
Figure
15:
Top
Level
of
the
TOGAF
Content
Metamodel.
For
more
de- tails
see
TOGAF
9,
figure
33-3.
http://www.opengroup.org/architecture/togaf9-doc/arch/Figures/34_contentfwk5.png
This is about where we will end with this book and this topic. It is not our aim to duplicate TOGAF texts in other words. Here we can only recommend referring to TOGAF chapters 33 thru 37 in order to get an idea what the Architecture Content Framework can offer you in case you need a Meta model as a basis to start storing information about your architecture.
7.2
Most organizations sooner or later arrive at a point where they want to put all relevant information from their enterprise architecture efforts into some form of repository and where they want to edit the viewpoints of their architectures in a single architecture tool. Lets have a look what TOGAF has to offer you, if you want to select a tool. There are two chapters, which are focused on the repository and tool matter:
44
Chapter 41 (Architecture Repository): Here you get 7 pages (pages 559 thru 565) of information of what TOGAF thinks should be in an Architecture Repository. There are references to the TOGAF ADM (Architecture Development Method) which as we have already demonstrated does not really cover strategic IT planning. As a consequence, the recommended content listed in TOGAF Chapter 41 also does not cover much of strategic IT planning. Chapter 42 (Tools for Architecture Development): gives you a 4-page list of evaluation criteria for architecture tools. If you have ever done a similar tool evaluation you know that criteria cata- logs for tool evaluations are way longer in most practical cases. And again the list does not cover strategic IT planning. Therefore, if you plan to acquire an enterprise architecture manage- ment tool have a look at the evaluation done by Technical University of Munich in 2008 (you find pointer to the material at http://wwwmatthes.in.tum.de/wikis/sebis/eamts2008). The study has a broader approach from the functional side. It also contains aspects of strategic IT planning. And the list of criteria used is by far more extensive than the one provided by TOGAF chapter 42.
7.3
Another
thing
one
would
like
to
find
in
an
Architecture
Framework
is
help
with
setting
up
an
Enterprise
Architecture
Management
Opera- tion
(or
call
it
the
EAM
department,
or
processes
if
you
want
to
be
more
modern).
TOGAF
offers
you
a
so-called
Architecture
Capability
Frame- work.
This
Framework
lists
a
set
of
instances
you
need
for
a
success- ful
EAM
operation.
And
again:
Strategic
IT
planning
is
not
in
the
cen- ter
of
it
so
you
will
not
read
anything
about
IT
Strategy
or
IT
portfo- lio
management
but
you
will
get
advice
on
how
to
use
the
TOGAF
ADM
to
develop
e.g.
the
systems
support
for
the
architecture
practice.
What
you
will
find
is
advice
on:
Establishing
an
Architecture
Capability
(Chapter
46):
This
tells
you
how
to
apply
the
ADM
to
an
Information
Systems
and
Technology
Architecture
not
the
one
for
your
company
but
for
the
architecture
practice.
It
is
more
or
less
a
list
of
action
items.
If
you
look
at
this
from
a
management
perspective
then
tool
sup- port
comes
very
late
in
establishing
a
successful
architecture
2009 Wolfgang W. Keller all rights reserved
45
practice. Hence the order of things in TOGAF chapters 45 thru 52 is debatable. Architecture Board (Chapter 47): Describes what an Architec- ture Board does and how to set it up (4 pages all in all). Architecture Compliance (Chapter 48): Contains a checklist for architecture reviews how to plan them and what to check. Architecture Contracts (Chapter 49): Gives a few hints on how to make agreements on architecture between the enterprise that sources some piece of software and the contractors. Architecture Governance (Chapter 50): Gives hints on how to enforce the enacted architecture guidelines. Has been discussed in more extent above in Chapter 6 of this book. Architecture Maturity Models (Chapter 51): Applies the idea of CMMI and other Capability Maturity Models to TOGAF. Rather brief and already referring to future editions. Architecture Skills Framework (Chapter 52): Defines a set of roles and skills needed in the opinion of the creators of TOGAF to fill out the roles.
Browsing through this material you might conclude that TOGAF has a somewhat technology- and model-centric view of enterprise architec- ture. The chapters start with establishing the right tool support for the Enterprise Architecture Practice; you get some information on Governance and Architecture Governance Bodies but you do not get a business driven immersion path that pushes you towards maximiz- ing business benefit from day zero.
46
This chapter will explain the parts of TOGAF that have not yet been mentioned. You will be given a short summary for each item.
8.1
Before we waste time explaining what the subject matter of the TRM is about, we better use one picture (Figure 16), which explains the subject area much faster than a verbal definition. TRM is about the services any technology stack needs to offer.
47
Figure
16:
Simplified
view
upon
the
TOGAF
TRM
Overview.
For
the
analogous
figure
in
TOGAF
9
see
Figure
43-2.
http://www.opengroup.org/architecture/togaf9-doc/arch/Figures/43_trm_detail.png
As in the III-RM (see next section) the main use of such a reference model is terminology and checklists. You dont argue about the terms and you can check for yourself whether you need the services enu- merated in the reference model and whether you have all services you need present in your technology stack.
8.2
The III-RM gives you a 20 page taxonomy for all kinds of application software in an enterprise that has the goal of so called boundary-less information flow, meaning an enterprise which is integrated with the outside worlds logistics chains via buying and selling circles. The benefit of the chapter is that you get a vocabulary for talking about enterprises integrated with the outside world via electronic data exchange. We will not reprint any Open Group drawings here. Mainly for copy- right reasons. Best have a short look at TOGAF Chapter 44
48
9.1
Throughout this book we have demonstrated that TOGAF is a very useful collection of many methods and tools that an enterprise archi- tect might need for his work. Nevertheless we could also see, that TOGAF is more a collection than a uniform planned piece of work. The Architecture Development Method is somewhat closed in itself so is the Content Framework. But many other things are collections of tools and additions to the ADM: Which is NOT negative! But as we assumed in the introduction and could now kind of prove: TOGAF is not an Enterprise Architec- ture for Dummies Cookbook. TOGAF has a clear focus on developing architectures: Be it single systems or subsystems, be it clusters of applications systems or be it a blueprint for the top level architecture of an enterprise. The last of these will hardly ever be designed on a drawing board but will be a result of organic growth plus either an explicit or implicit standard for an industry.
9.2
The two areas in TOGAF that make the kernel of the framework are: ADM: The Architecture Development Method is a really mature and extensive method, checklist and guide that you can employ when you want to design any part of your IT architecture. Architecture Content Framework: This Framework lets you bring order to your architectural artifacts even if it does not tell you how to create all of them. The meta-model part gives you an extensive library of meta-objects that you can at least consider using when you have to construct the specific meta-model that is able to answer the questions your management has for their En- terprise Architects in the context of their enterprise.
49
Besides that we already listed a lot of other useful stuff that you can find in TOGAF.
9.3
TOGAF Certification
If you study job adverts, theres a recent tendency to mention TOGAF certification whenever an Enterprise Architect is wanted. In this book we have demonstrated that TOGAF is not really strong at strate- gic Enterprise Architecture Management. TOGAF is strong at project architectures of all scales and some of the tasks of strategic Enter- prise Architecture Management. As with any certification, TOGAF certification is for sure benefi- cial to those who offer the courses and exams. If it comes to HR de- partments and senior management, the ones who often hire enter- prise architects, it is doubtful whether they have deep knowledge of the differences between Strategic IT Management and the kind of Enterprise Architecture that the TOGAF ADM is good at. So if you want to hire somebody who is able to perform Enterprise Architec- ture Management a TOGAF certificate covers maybe 30% percent of the job and to the authors regret the less important 30% because the important part is about business strategies and business man- agement. So for many Enterprise Architects the TOGAF certificate will be just another drivers license. The license says that you are entitled to drive it does not say that you are a good driver. Nevertheless you have to take the test if the public (represented by the government) makes it mandatory to have the license. Or if the majority of compa- nies require a TOGAF certificate people will need one as an entry ticket for certain jobs.
50
9.4
Future
Predictions are hard to make and making predictions by looking at the past often does not make too much sense. The first version of the TOGAF ADM was derived from TAFIM 2.0 around 1995. Todays ADM has a long history and no roots in strategic architecture management. In 2009 the Architecture Content Framework was added from the Capgemini IAF (Integrated Architecture Framework). This was made possible because Capgemini invested a lot in sup- porting the OpenGroup and TOGAF. Whether TOGAF will become a full framework for the more stra- tegic parts of Enterprise Architecture Management is hard to predict if not unpredictable. In any case it is very useful for developing architectures in concrete projects no matter whether they have a small or a very large scope.
10 References
51
10 References
[Alexander79a] Christopher Alexander: The Timeless Way of Building. Oxford University Press, New York, 1979. [Alexander79b] Christopher Alexander: A Pattern Language. Oxford Uni- versity Press, New York, 1979. [Bass+98] Len Bass, Paul Clements, Rick Kazman: Software Architecture in Practice. Addison-Wesley, 1998. [Broadbent+05] Marianne Broadbent, Ellen S. Kitzis: The New CIO Leader. Harvard Business School Press, 2005. [Brooks75] Frederick P. Brooks, Jr.: The Mythical Man-Month. Essays on Software Engineering. Rea-ding, Mass. et al.: Addison-Wesley 1975. [Buschmann+96] Frank Buschmann, Regine Meunier, Hans Rohnert, Pe- ter Sommerlad, Michael Stal: Pattern Oriented Software Architec- ture, A System of Patterns, Wiley 1996. [Clausewitz98] Carl von Claussewitz: Vom Kriege. Ullstein Taschenbuch 1998. [Coplien96] Jim Coplien: Software Patterns. SIGS Publishing, New York et al. 1996. [Coplien97] Jim Coplien: A Pattern Language for Writers Workshops; available at http://www.riehle.org/community-service/hillside- group/europlop-1997/p2final.pdf (link checked 2009/03/25) [Dern09] Gernot Dern: Management von IT-Architekturen. 3rd Edition, Vieweg Verlag, Edition CIO, 2009. [Gamma+95] Erich Gamma, Richard Helm, Ralph Johnson, John Vlissides: Design Patterns, Elements of Reusable Object-oriented Software, Addison Wesley 1995. [GarlanShaw96] David Garlan, Mary Shaw: Software Architecture: Per- spectives of an Emerging Discipline; Prentice Hall 1996.
10 References
[Hanschke08]
Inge
Hanschke::
Strategisches
Management
der
IT- Landschaft.
Ein
praktischer
Leitfaden
fr
das
Enterprise
Architec- ture
Management;
Hanser
Verlag
2008.
52
[ITIL02] OGC; Office of Governmernt Commerce: Best Practice for ICT In- frastructure Management. Published by TSO The Stationery Office, 2002. [Kaplan05] Jeffrey Kaplan: Strategic IT Portfolio Managment; Governing Enterprise Transformation; Pittiglio Rabin Todd & McGrath (PRTM) Publishing 2005. [Keller06] Wolfgang Keller: IT-Unternehmensarchitektur, dpunkt Verlag 2006. [Longepe03] Christophe Longepe: The Enterprise Architecture It Project: The Urbanisation Paradigm; Kogan Page Science; 2003 [Riege05] Christian Riege: Methoden fr das Management von An- wendungsportfolios eine vergleichende Untersuchung. Diplomar- beit, Universitt Leipzig, 2005. [Ross+06] Jeanne Ross, Peter Weil, David C. Robertson: Enterprise Archi- tecture as Strategy: Creating a Foundation for Business Execution. Mcgraw-Hill Professional 2006 [Schekkerman04] Jaap Schekkerman: How to Survive in the Jungle of En- terprise Architecture Frameworks. Verlag Trafford, Victoria, Canada, 2004. [sebis08] sebis: Enterprise Architecture Management Tool Survey 2008. Available from Technical University Munich, Chair of Florian Mat- thes; http://wwwmatthes.in.tum.de/wikis/sebis/eampc (link chec- ked 2009/03/25) [Shaw+96] Mary Shaw, David Garlan: Software Architecture Perspec- tives on an Emerging Discipline. Prentice Hall, 1996. [Sowa+92] J.F. Sowa, J. A. Zachman: Extending and Formalizing the Framework for Information Systems Architecture. IBM Systems Journal, Volume 31, No. 3, 1992. IBM Publication G321-5488. [Starke05] Gernot Starke: Effektive Softwarearchitekturen. 3. Erweiterte Auflage, Hanser Verlag, 2008. [TOGAF8.1.1] The Open Group: TOGAF Enterprise Edition. Version 8.1.1: available at http://www.opengroup.org/architecture/togaf8- doc/arch/. The Open Group, 2007. [TOGAF9] The Open Group: TOGAF Version 9; available at http://www.opengroup.org/architecture/togaf9-doc/arch/ ; The Open Group, 2009. 2009 Wolfgang W. Keller all rights reserved
10 References
53
[Ward+97] John Ward, Pat Griffiths: Strategic Planning for Information Systems. 2. Auflage, Wiley, 1997. [Ward+02] John Ward, Joe Peppard: Strategic Planning for Information Systems. 3. Auflage, Wiley, 2002. [Weill03] Peter Weill: Dont Just Lead, Govern! Vortrag DiamondEx- change, Juli 2003, http://exchange.diamondcluster.com/archives/200307/weill_0703. pdf (aufgerufen 2.1.2006). [Weill+04] Peter Weill, Jeanne W. Ross: IT Governance How Top Per- formers Manage IT Decision Rights for Superior Results. Harvard Business School Press, 2004. [WinFis06] Robert Winter, Ronny Fischer: Essential Layers, Artifacts, and Dependencies of Enterprise Architecture. In. Los Alamitos, CA, USA : IEEE Computer Society, 2006.- EDOC Workshop on Trends in Enterprise Architecture Research (TEAR 2006) within The Tenth IEEE International EDOC Conference (EDOC 2006).- Hong Kong.- ISBN 0-7695-2743-4, page 30. [Zachman87] John A. Zachman: A Framework for Information Systems Architecture. IBM Systems Journal, Volume 26, No. 3, 1987. IBM Publication G321-5298. Can be obtained via http://www.research.ibm.com/journal/sj/263/ibmsj2603E.pdf (Link checked am 2005-01-27). [Zachman97] John A. Zachman: Enterprise Architecture: The Issue of the Century. Database Programming and Design, March 1997.