 |
 |
 |
 |
 |
Acronym for enterprise a pplication i ntegration .
EAI is the unrestricted sharing of data and business processes throughout the networked applications or data sources in an organization.
Early software programs in areas such as inventory control, human resources, sales automation and database management were designed to run independently, with no interaction between the systems.
They were custom built in the technology of the day for a specific need being addressed and were often proprietary systems.
As enterprises grow and recognize the need for their information and applications to have the ability to be transferred across and shared between systems, companies are investing in EAI in order to streamline processes and keep all the elements of the enterprise interconnected.
Over the last ten years, the business sector made unprecedented investments in information technology.
Two distinct developments drove the need for these investments:
The introduction of enterprise framework applications, and the emergence of the World Wide Web, e-mail,
and Internet applications.
Enterprise framework applications that were designed to reengineer core business operations include a wide range of applications including Supply Chain Management (SCM), Enterprise Resource Planning (ERP),
and Customer Relationship Management (CRM).
These sophisticated applications required substantial capital, specialized technical resources,
and operational restructuring to implement.
Companies that successfully deployed these enterprise frameworks derived significant efficiencies
in their core operations that in turn translated into a competitive advantage.
Web applications and e-mail, which were typically designed for messaging and information exchange,
were based on open standards and proved relatively easy to implement.
Their deployment produced new capabilities and efficiencies for communicating information that resulted in improved workplace responsiveness and performance.
In companies everywhere, the proliferation of enterprise applications was accompanied by a simultaneous build-out of a computing and networking infrastructure designed to facilitate information exchange.
The complications that have emerged from this proliferation of diverse technologies are due not to the multiplicity of the systems themselves, but rather to the strain on programming resources and IT budgets to make information residing in these application silos accessible and usable to other platforms, enterprise employees, and partners and customers.
With the widespread information processing and communication capabilities found in organizations today, the demand for information has become overwhelming.
Every information worker with an Internet-enabled computer has the ability to access boundless information and computing functionality, though not necessarily the information or functionality that is most relevant to their day-to-day business tasks.
The growing disparity between what we expect our information technology to provide, and what it actually delivers is the primary reason that enterprise application integration (EAI) and business process automation (BPA) projects are the number one IT priority in most organizations.
The problem is that enterprise framework applications consist of thousands of program modules, databases, and data files with operational procedures, controls, and access mechanisms that are extensive and rigid. Developing extended programmatic capabilities or attempting to make information accessible in ways that are not defined in these systems requires enormous resources, time, and capital because the work involved consists of numerous sequential, low-level programming tasks.
Hard-coding point-to-point integrations, is the prevailing method for information exchange. Programmers knowledgeable in the APIs of the interfacing applications develop custom programs to access a source application ' s data (usually in binary format); map and convert the respective data structures;
manipulate the data as required; and deliver it into the target application.
This produces a tightly coupled, highly specific set of functions that exist and execute in the form of procedural code, just like the applications themselves.
This type of development effort is highly linear;
each step is dependent upon the completion of a previous step and cannot be broken out easily, or at all, into independent tasks executed by distributed resources. Consequently, scaling to meet a growing workload of integration projects means adding more programming resources.
The extent to which integration projects consume resources can be expressed by the N-Square equation: N* (N-1)/2, where N is the number of interface endpoints. If an organization has a fully meshed distribution matrix of just 20 inter-exchange endpoints (a very low number), 190 programmatic inter-exchanges must be developed. Because each integration interface is specialized and manifested in an encoded construct that is not modularly reusable, overall programming efficiency is not maximized by the proliferation of programming resources. As more integration requirements arise, they continue to overwhelm IT, eating up both resources and budgets. Not surprisingly, then, in most organizations, functions that require an automated solution continue to be executed on a manual basis.
Both the software development community and end users alike understand that a major impediment to using information technology more effectively throughout the enterprise is the linear and costly process of making information available and usable to multiple applications and processes. This impediment prevents businesses from being able to create agile, process-centric business environments that can organize, monitor, and modulate themselves to achieve operational equilibrium in response to both subtle and gross changes in the business environment.
Fortunately, a new computing paradigm has emerged that alleviates the inefficiencies of EAI and BPA development, and the software standards bodies have worked quickly to codify its protocol methodologies. The defining concept of this paradigm is the elevation of the integration process from the program layer to the information (document) and transport (messaging) layer. By separating information from the applications that use it, exposing it as clear text, and using self-describing XML metadata to give it meaning and structure, the information can be processed by any application capable of parsing and interpreting XML metadata. Even the operational functions and invocation methods of applications themselves can be described and exposed using XML, allowing them to be executed without regard for where they reside, how they were originally developed, or what platform they run on. This is the underlying premise of the Web Services protocols, Simple Object Access Protocol (SOAP), and Web Services Definition Language (WSDL).
