Systems dynamics study of integration in information technology

Systems Dynamics
Study of Integration in
IT

Information Technology offers a continuous choice of
options to be harnessed and used to support business. IT
applications have specific domain focus to solve business
problems. Since business is always dynamic the
technologies offered to support business are always
being improved to promote flexibility agility and user
friendliness. Integration domain helps in streamlined
integration of various enterprise applications, data,
business process etc. Applications, technology and
software belonging to this domain are perhaps the most
dynamic of all other IT domains. The paper gives a high
level view of technology lifecycle of various integration
lifestyles. Understanding the various forces acting when
a technology is being offered will help an IT worker,
whether he is an architect ( for choice of integration
style) or a project manager ( for investment decisions in
specific technology) or an IT consultant ( for decisions on
learning specific technology)

1 Introduction
All applications in the organization are developed in a heterogeneous environment.
There is a constant need to share information between various applications across the IT
landscape. Hence integration between these applications are important. Software tools
developed are trying to facilitate development efforts in integration and as such there is a
constant product innovation in the market. When new products are launched the reaction
from the firms are of varied degree and it can be anything from the following: early
adoption, late adoption, sheer rejection. Since there are multiple stakeholders in this
market the point of views put forward often have conflict of interest.
The focus of this paper would be to develop a framework to understand the lifecycle of
the Integration style with special focus to SOA. The model tries to find the drivers for
emergence of an integration style with various data related to SOA1 , lifespan of SOA and
time required for emergence of new area such as Integration-As-a-Service2.

2 Objective

• Build a systems dynamics model for development of integration style
• Identify Drivers for development of Integration Technology
• Predict technology lifecycle of the current Integration Style
• Predict Adoption of Alternate Integration Style
• Use the framework to decide on new technologies available

3 The Model
3.1 Problem Articulation
Integration Domain has been in constant flux with the single theme of integration
of business data, process or application. New requirements within business are always
driving new application/software development. Applications business logic is acted on
the data and information which often comes from other applications. For example a CRM
solution being built may need data from an already existing ERP application. This would
need an information flow from ERP application to CRM application and hence there is a
need for integration.

3.1.1.1 Integration Spaghetti
Integration, done without any foresight often leads to a situation which has been
referred to as Spaghetti Integration3 in many journals and articles. IT skilled workers
have always been offered myriads of technology options for integration. With such a
wide range of options and no focused strategic use of integration inappropriate use of
technology is quite common. At any point of time an Integration architect has the
following options:

2|Page

i. To develop a quick fix solution with existing tools
ii. To develop a strategic solution with existing tools.
iii. To invest in new integration tools for better handling the integration
requirements.

3.1.1.2 Architect’s Dilemma
IT applications are normally considered part of operation and in many firms IT is
not considered strategic but a backroom operational process to optimize costs. Even if IT
application developed has strategic business significance Integration is one domain where
strategic planning is virtually nonexistent. Thus an architect and project manager in many
case continue doing integration using old tools and old methods.
Integration is always in flux and technology offering from this space is quite
dynamic. So there is a tendency to look at new options as some kind of hype. This is not
uncommon there is often a hype associated with many technical innovations4 . But if an
architect always thinks of new Integration Style as hype there is a good chance that firm
may get delayed in adopting new integration technology which could have added
significant business value. At this stage the architect or the firm may completely miss the
improved technology wave or if there is no proper way to manage new technologies he
may bring in outsiders as consultants and depend on them. In both cases (missing the
wave and bringing in outsiders) has the chance of IT division within the firm losing out
on learning. When outsiders (consultants or contractors) are brought in it requires
substantial time to understand the organization culture and business. Thus it may take
some time before realizing business when outside experts are invited.
There is an option to continue as it is or invest in new technology. While investing
in new technology a farsighted IT manager or an architect focused on Integration should
also have an idea how long the integration framework or supported by a specific
commercial product proposed by the new technology would last. This is important as
with every investment there is always a question on ROI. Thus an IT Integration
manager will always remain in the dilemma which perhaps can be termed as Architect’s
Dilemma as he will always have the choice to pick form variety of design and technology
options.
Each and every domain has its own in inherent dynamic behavior and it affects the
IT applications in a similar fashion. For example marketing domain is much more
dynamic than domain of human resources. Hence there are chances that applications
supporting marketing will come up quickly compared to applications. For example we
have CRM applications, SalesForceDot Com where as HR nothing much is heard beyond
outsourcing of payment processes and generations of pay stub.

Nevertheless whatever happens in rest of the domains integration is one domain
which always remains under pressure to deliver and support business process. Since rest
of the business process is changing there is always an need to connect this various pieces
of business. Thus an architect in the domain of Integration will have more than fair share
of Architect’s Dilemma.

3|Page

3.2 Reference Mode of Behavior

3.2.1.1 Integration Style
Integration in the paper would mean technical options to share information across
various applications. There are various kinds of integration patterns. Integration style is a
combination on Architecture style implemented using a specific group of products. Thus
an SOA architectural style of integration can be implemented over SOAP/HTTP
framework provided by any tool. Technology Lifecycle often follows growth pattern
similar to product lifecycle i.e. S shaped Curve. If specific integration technologies are
considered from past in most cases the curve seems to follow the S shaped curve.
Integration has various technology options for example RPC, Com D-Com, CORBA (in
the past) along with many architecture patterns like Hub and Spoke, SOA etc. Also on the
horizon there is a virtualization and Integration as a Service. For each architectural style
there will be associated products (commercial or open source) in the market. A firm
trying to leverage on a integration style will have to buy or get a off the shelf software
which will provide frameworks, features to implement the Integration solution. Thus an
integration style will be combination of architecture style and associated product
offerings. Within a firm thus there is a tight coupling between architecture styles a
product supports.

3.2.1.2 Reference Mode
If we have to think of a reference mode of integration style we have to study the
reference mode of its constituent variables namely Architectural pattern ( e.g. SOA),
Product/software( e.g. Mule) and technology ( e.g. Soap,HTTP). The reference mode of
each of these variables will be somewhat be similar to a Technology Lifecycle. Thus
product lifecycle and technology lifecycle has similar reference mode of behavior except
for the fact that a architectural life often outlives a product. For example SOA is an
architectural style and there are many products (Example WebSphere, BizTalk, Mule etc)
in the market which provides framework for SOA. Hence it cam be said the Product
lifecycle is smaller than architecture lifecycle.

Figure 1 Architecture Style Vs Product Lifecycle

4|Page

The blue curve is the architectural style and the green and orange curves are hypothetical
product lifecycle curve. The reference mode behavior is not drawn to scale.
Thus the Integration Style will also have a reference mode of behavior as follows:

Activity

Time

3.3 Formulating Dynamic Hypothesis

Much has been stated already in the paper to indicate that the technical world of
integration is quite dynamic. At the time of writing this paper there are quite a few
integration styles looming large in the horizon. Some of them are:
• SOA –SOAP-HTTP-WS*
• SOA-Rest over HTTP
• SOA-Others5
• SOA-BPM
• Integration as a Service.

It can be found out that SOA is an architectural style which can be implemented in
various ways and some of them have just been mentioned. There are different products in
the market and many of them provide SOA framework with partial to many of the
features or styles listed above. The focus of the paper is to have a high level view of
integration styles and predict which part of the lifecycle each style belong to. It also aims
to develop a framework for forecasting lifecycle of integration style. The following
hypotheses are being made:

i. SOA-SOAP-HTTP Integration style is currently at its peak
ii. Firms have to adopt a portfolio approach on integration style

5|Page

3.4 Causal Loop Diagram

ISFocus AltISFocus

ISAdoptionRate ISDesertionRate

IndustryCooperation
AltIsIndustryCooperation

ISAdoptionSense AltISAdoptionSense

ISAdoptionEase ISTechLimitation
AltISTechLimitation AltIsAdoptionEase

AltISMaturity
ISMaturity

TechnologyInnovation

IntegrationMarket

3.5 Variables

The Variables in the Model can be grouped in three categories: Variable for
current Integration Style, Variables for Alternate Integration Style and Generic
Variables Applicable to both style.

3.5.1 Group A( Generic Variables)

Integration Technology ISMaturity
Market Innovation

6|Page

I. Integration Market: Integration market opportunity. In an article Gartner indicates
around 30% of the 872454 Mn dollar IT services belongs to development and
integration 6. This huge market opportunity is one of the key driver for various IT
firms (software vendor, IT services providers) to continuously invest and
innovate.

II. Technology Innovation: This is a variable to indicate the rate of innovation. High
value of Technolgy Innovation indicates high product maturity and hence
Integration Style maturity

Technology Innovation = Integration market × k1
k1 is the coefficient to convert integration market opportunities to technology
innovation. This is a proxy parameter to indicate level of activities in terms of
R&D effort, new product developed ( e.g ESB)7, new architectural style proposed
(e.g REST-HTTP)

3.5.2 Group B (Variables Applicable for Current Integration Style)

Industry TechLimitation
Cooperation

+
Adoption Sense AdoptionEase

+
ISFocus ISMaturity

III. ISMaturityAge: This is a proxy variable indicating maturity of the Current
Integration Style. This measured by a number which is proportional to number of
years the Integration style has been prevalent in the industry.

IV. ISMaturity( Integration Style Maturity): Technology innovation influences the
growth of the IS Style maturity

ISStyleMaturity = Technology Innovation × ISMaturityAge × k2
7|Page

k2 is the coefficient to convert the Technology Innovation to appropriate
ISMaturity.

V. ISAvailableProducts: This is a dimensionless variable. For every integration style
available with there are usually a set of products available in the market. The
value for this has been arrived by weighted avg of the products. A higher weight
signifies feature rich product

ISAvailableProducts= [(∑ Product i × Weightage i ) ÷ n] where i= 1,2,..n

VI. Adoption Ease: This is a dimensionless variable. Products which provide options
to quickly develop and implement solutions will score high on Adoption Ease
scale. Features such as matured IDE or various configuration options or other
“cool features” often create ease of adoption. Products need to be at certain level
of maturity and complexity to provide various cool features to help in
development. For example IDEs like Eclipse has various options to declare define
variables and avoid errors. Commercial Middleware’s such as webSphere IBM or
SAP PI are much more complex and development and coding activities have been
reduced to only configuration. The adoption ease value for this software products
are quite high

Adoption Ease = ISMaturity × ISAvailable Products

VII. ISTechnicalLimitations: This is a dimensionless variable on a relative scale.
These are limitations for a certain integration style which arise because standards
are not clear or perhaps the developer needs to take care of many low level details
while implementing the solution. 8
TechLimitation = ISMaturity × (1÷ ISMaturity) × k5

k5 being the coefficient to convert ISmaturity to TechLimitation

VIII. Adoption Sense: This is a dimensionless variable. Adoption makes sense when it
satisfies three important criteria
a. Ease of Adoption (i.e. how developer friendly the product is) sense
indicates whether it makes sense to adopt the integration style.How
Reusable components and interfaces can be built using the Integration
style
b. Technical Limitation: Lesser the technical limitation the better sense it
makes for adoption

8|Page

c. Industry Cooperation: More the industry cooperation in terms of uniform
standards support by multiple vendors more it makes sense for the firms to
adopt a certain integration style.

There is no precise mathematical way to model the complex behavior to arrive at
these values. For simulation purpose the following equation has been considered

Adoption Sense= Adoption Ease × Market Cooperation – Technical limitation

IX. ISShareofIndutryCooperation: This is a dimensionless variable.
ISShareofIndutryCooperation is a fraction of the Total industry cooperation.
Total Industry Cooperation is a combination of many factors. Cooperation
initiatives are done actively and passively in many ways by the stakeholders in the
entire eco system. It starts from marketing and sales launched by product
companies to thought leadership by industry speakers, articles and blogs written
by independent consultants. This is a cooperation which works in a self sustained
manner. The fact that consultants and industry speakers will adopt a new
technology is not only driven by the notion that a new concept may help solve
integration issues and facilitate business but also because it will help them to be
establish credibility through thought leadership . Thus once the concept is adopted
by critical mass of people these thought leaders are suitably rewarded in economic
sense. 9
Industry cooperation also refers to cooperation between big firms on agreeing and
promoting common standards

X. Staff Maturity: This is a dimensionless variable. The success and failure of any
technology depends ultimately on the staff involved in implementing it. No matter
how promising a technology may seem to be its success ultimately depends on the
employees. Appreciation for new technology, ability to understand the merits of
different technologies, and application of different technologies for solving
different problems are some of the key traits of matured staff.

XI. IS Adoption Rate (Integration Style Adoption Rate): This is a variable to indicate
the integration style adoption rate within firms. This is a dimensionless variable

IS Adoption Rate=Adoption Sense × Staff Maturity × k3

k3 is the coefficient to convert staff maturity to appropriate ISAdoption rate

XII. ISFocus (Integration Style Lifecycle): Firms at any point of time will be focused
on specific integration style. For example throughout during 2005-2007
Integration vendors offered EAI tools like webMethods , TIBCO offered and
promoted hub and spoke style of integration using message brokers. This was not
only represented by the increasingly huge number of projects that author

9|Page

experienced as a consultant but also by the large number of articles published in
the magazines and articles. Thus, at the time of writing this article a significant
number of firms are focused on SOA style of integration. With time the focus will
gradually shift to different integration style. Thus for simulation purpose:

∫ d(ISFocus ) = ∫ ISAdoptionRate . dt - ∫ ISDesertionRate .dt

10 | P a g e

3.5.3 Group C (Variables Applicable for Alternate Integration Style)

Industry AltIS
Cooperation TechLimitation

+
AltIS AltIS
Adoption Sense AdoptionEase

+
Alt AltIS
ISFocus Maturity

XIII. ISDesertionRate ( Integration Style Desertion Rate). This is a variable to signify
the Desertion rate from current Integration style. A higher desertion rate style
with respect to the Adoption rate will signify the shift in trend to the new
Alternative Integration style. When desertion rate is high than adoption rate there
will be steady decrease in value of the ISFocus. ISDesertion rate is influenced by
alternative Integration Style Adoption Sense and Staff maturity. The equation
governing the nature of this variable occurs in the similar fashion as that of the
ISAdoption rate but works against the current integration style. Hence

ISDesertionRate = AltAdoptionSense × Staffmaturity

XIV. AltISAdoptionSense. This variable is similar to AdoptionSense but it signifies
how much sense it makes to adopt the new/alternate Integration Style.

AltISAdoption Sense = (AltISAdoptionEase × AltISShareIndustryCooperation )–
AltISTechLimitation

XV. AltISAdoptionEase: This variable is again similar in nature to ISadoption ease
except for that it works for new/alternate Integration Technology Style

AltISAdoptionEase=AltISMaturity × ALtISAvailableProducts

11 | P a g e

XVI. AltTechLimitation. Variable similar to ISLimitation

[AltISTechLimitation= 1÷ AltISmaturity]

XVII. AltISMaturity: Similar to ISMaturity except that technical limitation in one
product will influence growth of alternative Integration Style. This is quite
common in the industry as one Technology paves way for another technology as
it hits its limitations. Hence the equation is slightly different from ISMaturity.
AltISMaturity = AltISMaturityAge × TechnologyInnovation × k7 × k6

k6 is the coefficient which converts IS tech limitation to AltIS maturity
k7 is a variable similar to k3 and its values are similar to k3 except for that it
contributes to the AltMaturity after sometime

XVIII. AltISMaturityAge: Similar to ISMaturity Age

12 | P a g e

3.6 Loops Analysis
3.6.1 Reinforcing Loops of Integration Style
ISFocus
Maturity

ISMaturity
There are two reinforcing loops in this model one
ISAdoptionRate for Current Integration Style and Another for
alternate Integration style. With the increase in
Integration Style maturity there much more tools
and products in the market which ultimately
ISAdoptionEase ISAdoptionSense
increase the adoption rate. With the increase if
adoption rates the focus on the specific integration
style increases. As focus on the specific integration
style increases product companies and industry
work as a whole to improve the IS Maturity and this cycle keeps on repeating till it
reaches a saturated level of ISFocus.
Similarly the same cycle exists for Alternate Integration style. However both the
cycle don’t work at the same time. There would be a switch form one cycle to another
cycle with the passage of time. This switch form one cycle to another cycle takes effect
when one integration style hits saturation level or there is attractive alternate technology
options.

ISAdoptionRate AltISAdoptionRate

ISFocus AltISFocus

IS
ISAdoptionSense AltIS AltISAdoptionSense

AltIsMaturity AltISAdoptionEase
ISMaturity ISTechLimitation

K6
Switch

3.6.2 Reinforcing Loops of Industry Cooperation
There are reinforcing loops : one for the existing Integration Style and another for
the Alternate Integration Style.
Once the Integration Focus increases it influences the Industry which helps the
promotion of the integration style. Integration Service Providers , Software Vendors,
Independent Consultant, Industry Though Leaders , IT Consultants all ride in the wave of
the new technology style. This is a very self sustained reinforcing model. All IT
13 | P a g e

stakeholders participate in promoting the integration style which influences the IT firms
to adopt the Integration style. Architects in the firm when looking for information on the
integration style consult the independent consultants or white papers , all of them has the
same information which promotes the integration style. This facilitates adoption within
the firm and thus the cycle keeps on going.
A similar reinforcing loop exist for Alternate integration style

3.7 Running the Model
In this model Vensim 5.9 is being used for simulating the model and investigating
various key impact factors. It has been simulated for 20 years starting from year
2000.

3.7.1 Parameter Values

The model has been run with two different scenarios one the current SOA-HTTP-
SOAP style and another Integration As a Service. There are similar set of variables for
each technology scenario. The values for each of these variables are different.

3.7.1.1 Independent Variables
There are some independent variables like ISAvailableProducts, ISMaturityAge,
AltISProductsAvailable, AltISMaturityAge, k1, k1 k3, k4, k7. The values for theses
variables and their justification is in the following tables

Variable Equation Values Explanation
ISAvailableProducts (∑ Product i × Weightage i 3.06 All technology stack like
)÷n Fusion(Oracle),
where i= 1,2,..n webSphere(IBM), PI ( SAP),
BizTalk ( Microsoft and host
of other open source ESB
products were considered.
AltISAvailableProducts (∑ Product i × Weightage i 1.2 To calculate the value of the
)÷n AltISProducts products
where i= 1,2,..n supporting Integration-as-a
service, products from
CastIron10 and few other
firms were considered.

ISShareofIndutryCooperation NA 0.42 ( at pd =1) Industry Cooperation to
promote certain integration
style. Proxy parameter
considered has been search
results in using google search
engine11 . With time the value
changes. Values are
calculated based on this :
(0.42+RAMP(0.2, 1 , 14 ))
AltISShareIndustryCooperation NA 0.29 Similar to
ISShareofIndustryCoperation.
Values are calculated based

14 | P a g e

Variable Equation Values Explanation
on this :
(0.42+RAMP(0.2, 1 , 14 )).
ISMaturityAge
Staff Maturity Constant 5 This is a proxy number being
modeled to reflect the staff
maturity
k1 NA 0.17 Market to Innovation
Coefficient. The value of this
variable is taken from the
R&D budget percentage of
TIBCO,12 a popular
Integration product.
k2 NA i = 1,2,…12 Maturity to Innovation
coefficient for existing
Integration style. k7. Initial
Value of this is 2 and for each
unit of time it increases by 1
till it reaches a saturation
point 12 at year 12.
k7 NA i=0.7 (for pd 1- Maturity to Innovation
8, coefficient for Alternate
Integration Style.
0.8, 0.9, 1, 1.2, The values of k7 would be
1.3, =1.4(for pd
15-18)
quite low compared to k2 as
its initial years Alternate
Integration style
k3 NA 1 AdoptionSense To
AdoptionRate Coefficient
k4 NA 1 AltAdoptionSense To
ISDesertionRate
k5 NA 30 Maturity to TechLimitation
Coefficient
k6 10/ISTechLimitation NA Coefficient for
converting Technical
limitation of one integration
style to promote Maturity of
alternative technical style

3.7.1.2 Dependent variables
There are the dependent variables which are derived from the independent variables
through various mathematical formulas. The dependent variables along with their
supporting Vensim equations has already been explained in Section 3.5

15 | P a g e

3.7.2 Simulation Run

ISFocus
60,000

30,000
Dmnl

0

-30,000

-60,000
0 2 4 6 8 10 12 14 16 18 20
Time (year)
ISFocus : Current
ISFocus is an indication of activities around certain Integration Style. As proposed in the
beginning of the paper [see Reference Mode of Behavior] that any Integration Style will
follow a curve similar to Product Lifecycle which has been validated through the curve.
ISFocus is a Level variable and its behaviors is controlled by two variables one working
as inflow and other as outflow:

ISAdoptionRate
20,000 ISDesertionRate
40,000

15,000
30,000
1/year

10,000
1/year

20,000

5,000
10,000

0 0
0 2 4 6 8 10 12 14 16 18 20 0 2 4 6 8 10 12 14 16 18 20
Time (year) Time (year)
ISAdoptionRate : Current ISDesertionRate : Current

16 | P a g e

3.7.3 Implications
Some of the variables play more important role as they directly change the behavior
of the system. Out of all the variable the independent variable are key to the
changing the input pattern of the model. The significant variables are mentioned
below:

i. Industry Share of Cooperation: This variable directly controls the
adoption sense for any integration style. The value of this variable
doesn’t change quickly with time. Industry cooperation is a variable
which is influenced by many stakeholders. There are some important
stakeholders who can change the direction of the technology (see note
9).
ii. k6 Switch (Coefficient for converting Technical Limitation of one
integration style to promote alternative Integration Style) :
The trigger to shift form current integration style to another integration
style is the technical limitation of one Integration style. It should be
understood that limitations in one technology paves way for other
technologies. Two integration style has been evaluated in this paper : In
house SOA-HTTP-SOAP style and Integration-As-As Service. Both are
completely different models hence a complete switch from one form to
another form is not expected. Integration As a Service promotes
virtualization of integration efforts. This will mostly happen because of
limitations of in house SOA-HTTP-SOAP style. In spite of much
promise and awareness on this style it still is not as flexible and agile as
business would like an integration development want it to be like. With
virtualization of Integration efforts all non core integration efforts, B2B
integration efforts would clearly get outsourced to this new integration
style.
iii. Lifecycle of Integration Lifecycle:
The model was simulated for 20 years with two Integration style in
focus. Although the lifecycle of Integration may seem to last for 15
years the peak lifecycle for any Integration style is only 5-7 years. This
is also in agreement with the experience by Integration consultants in
the IT firms.

17 | P a g e

3.7.4 Limitations of the Model

a. Industry Cooperation:
Industry Cooperation is a function of many variables

Total Industry cooperation = ISShareofIndutryCooperation +
AltISShareIndustryCooperation + Other Industry efforts.
Increase in share of industry cooperation in alternate Integration style will be
marked by decreased of Industry cooperation of current integration style. This
dynamic behavior of the two variables have not been captured in the model

b. Influence on ISMaturity:
An increased focus on Integration Style ( i.e increase in value of ISFocus)
influences maturity of the Integration style till it reaches a stage where the
limitations cannot be removed anymore and new Integration Style needs to be
adopted. The feedback connection between ISFocus and IS maturity as not been
captured in the simulated model.

c. Multiple Integration Styles:
Although the model has been simulated with two integration style but at any
given point there may be multiple Integration Style option available. Different
kinds of limitation may lead to different kinds of integration. For example
limitation of SOA supporting products frameworks may be heavy investment
which can promote Integration-As-a Service Style of Integration where as
limitation SOAP messages may promote HTTP-REST architectural style. Thus the
Integration Style SOA-SOAP-HTTP is promoting two alternate style of Integration.
This dynamic behavior of various integration styles has not been captured in the
model

3.8 Conclusion
In last 35 years Integration efforts has seen wide range of options from RPC,
Webservices to much as sophisticated BPM modeling tools. The technical options for
Integration domain will continue to evolve.
While an Integration lifecycle starting with initial conception may be around 15
years, Integration Style peak lifespan of any integration style is only 5-7 years. Since
change is inevitable firms should be careful to evaluate whether an integration style
available will support the strategic direction of business and hence IT. Firms should also
carefully scan their own integration technology landscape to identify and retire legacy
technology and integration options which pose risk for IT operations. A certain
Integration Style involves investment and commitment for the IT department. But
investment commitment doesn’t guarantee that newer and better Integration styles won’t
evolve. Firms shying away from newer and better technical options may lose opportunity
18 | P a g e

to learn as well as option to leverage IT for significant business productivity. Thus at any
given point of time firms should manage various Integration styles as a portfolio of
options.

19 | P a g e

1
SOA= Service Oriented Architecture
2
Integration as a service: Integration-as-a-service is the ability to deliver a complete integration stack from
the cloud, including interfacing with applications, semantic mediation, flow control, and integration design.
In essence, integration-as-a-service includes most of the features and functions found within traditional EAI
(enterprise application integration) technology but delivered as a service.
Source: Linthicum,David S."Defining Clouds For Enterprise" in Cloud Computing and SOA Convergence
in Your Enterprise: A Step-by-Step Guide

3
Spaghetti Integration: is a process of integration of the systems where each system is interconnected to
each of the remaining subsystems. When observed from the perspective of the subsystem which is being
integrated, the connections are reminiscent of a star, but when the overall diagram of the system is
presented, the connections look like spaghetti, hence the name of this method. The cost varies due to the
interfaces which subsystems are exporting. In a case where the subsystems are exporting heterogeneous or
proprietary interfaces, the integration cost can substantially rise. Time and costs needed to integrate the
systems increase exponentially when adding additional subsystems. From the feature perspective, this
method often seems preferable, due to the extreme flexibility of the reuse of functionality. Source:
http://en.wikipedia.org/wiki/System_integration#Methods_of_integration

4
Gartner Publishes Hype Cycle for emerging technologies every year. For the curious mind here is a the
link to 2009 Hype Cycle 2009. http://www.gartner.com/it/page.jsp?id=1124212
5
SOA can be implemented over many technologies. Others in general refer to other options available.
6
http://www.gartner.com/it/page.jsp?id=728811
7
ESB : http://en.wikipedia.org/wiki/Enterprise_service_bus
8
An example of product limitation could be features absent to implement a single-sign-on. For simulation
purpose it has been considered that TechLimitation is inversely proportional to ISMaturity which implies
that more the IS style is matured technical limitation would be less till it reaches a saturation level.
9
It’s a common knowledge that in industry there are various kinds of forces starting from
competing vendors developing products in same platform like Java, to vendors who develops products in
different platforms like .Net and Java. It won’t be a wrong statement to make if we state that firms will
adopt or promote a certain technology or integration style only if it suits their business. For example during
the times of CORBA there were two different competing technologies like CORBA (promoted by OMG)
and Com-DCOM(promoted by Microsoft ) was competing against each other . In the end although the
technology was promising it was dropped.
Another common example is HTTP. Most of the Internet traffic is based on this protocol and its
simple to adopt. Thus there is a virtual cycle around the use of HTTP. This is important as SOA-SOAP-
HTTP-WS* integration style will have major industry cooperation since products will definitely support
HTTP and XML. For example an architectural style (like SOA etc) , Protocol ( like TCP, HTTP etc)
message formats (SOAP which is based on XML) etc. Industry co-operation depends on to what extent
each of these components is supported by the industry vendors. At the time of writing this article there
seems to be wide consensus for this integration style (SOA-HTTP-SOAP-WS*) except for a group which is
favoring adoption of SOA-HTTP-REST integration style
10
www.CastIron.com

20 | P a g e

11
Industry cooperation can often be in terms of blogs, articles, etc. Ideally each of these events are
reflected by creation of webpage. As such a search engine like GOOGLE can easily find in on web. Hence
a proxy parameter has been considered to find out the industry cooperation. Its being assumed that a
Integration style will promote more responses if industry cooperation is high. Google search on following
terms reveals the following :
Term SOA= 37,500,000 results
CORBA= 3,650,000 results.
Thus various terms like Enterprise Application Integration, SOA, CORBA, Integration As a Service,
SOAP-HTTP, SOA –REST were considered. So to calculate the values for an Integration Style following
formula was considered:
[(Search results For SOA Specific terms)÷(Total Search Results For all key terms Related to Integration)]
Values were calculated using the following tables:
Term Results
CORBA 3,650,000
SOA 37,500,000
Enterprise Application Integration 6,460,000
Cloud Computing 28,000,000
Integration-as-a-Service 32,100,000
SOAP HTTP 13,500,000
SOA REST 4,230,000
Cloud integration 7,230,000
Total 121,210,000

Thus for the term ISShareofIndutryCooperation=( SOA Search results+ SOAP HTTP Serach
Results)/ Total Search Results
12
www.tibco.com

21 | P a g e

Systems dynamics study of integration in information technology

More Related Content

What's hot (20)

Viewers also liked (15)

Similar to Systems dynamics study of integration in information technology (20)

Systems dynamics study of integration in information technology