Patrick Casimir

MMIS 620: Management Information Systems

Assignment 6: Internet Research & Development Paper - Set A

Table of Content

Topic 1

Unique Title 1

Statement of the Problem 1

Outline of Paper 2

Reference List 2

Topic: 25. Enterprise Application Integration (EAI) over the Internet

Title: Enterprise Application Integration: Obstacles and benefits of implementation.

Statement of the Problem:

The implementation of enterprise application integration (EAI) has allowed companies to realize increasingly complex business processes. Some of the popular include web-based

Commerce, business-to-business commerce, customer self service, expanding or integrating enterprise resource planning (ERP) systems, mergers or acquisitions, compliance with major regulatory changes, business intelligence initiatives, supply chain integration and mobile computing. Yet the costs of implementing EAI can also be substantial, both from a financial standpoint and in terms of the organizational disruptions that are often involved. According to Paul Korzeniowski, “While EAI tools have many attractive features, there are implementation challenges. For instance, moving from a fragmented to a centralized buying system is difficult” (Korzeniowski, 2001, p. 120). Cheryl Kridva wrote in the EAI Journal : “For sites integrating custom-developed applications, the benefits of implementing EAI are impressive. But, the implementation process can be very time-consuming and costly up front ” (Kridva, 1999, p. 224).

Paper Outline

I. Introduction

A. What is EAI?

B. Where did EAI come from?

II. Why companies are implementing EAI?

A. Ability to integrate applications

B. Infrastructure adaptability

C. Support for multiple integration topologies.

D. Power to handle complexity

III. EAI structure

A. Business Process Integration

1. Process of workflow modeling

B. Application Integration

1. Data translation and transformation

C. Component Integration

1. Application servers

D. Data Integration

1. Tools for extracting, transforming, and loading data

E. Platform Integration

1. Messaging

IV. Challenges in implementing EAI

A. Integration Approaches

B. The Learning Process Associated with EAI

C. Customization

D. Producer’s and consumer’s integration

E. Cost

V. Solutions

A. Consulting and in depth evaluation

B. Training sessions

C. Strategic development

D. Effective data analysis

C. Ease-of-use incentives

VI. Future trends

A. Adding Process Integration and Monitoring Capabilities

B. Consolidation of Different EAI Applications

C. Implementation of EAI applications into new markets

VII. Conclusion

Reference List

Korzeniowski, P. (2001). EAI tools tie business to business. EAI Journal, 3, 118 – 125.

Kridva, C. (1999). Enterprise application integration. EAI Journal, 1, 223 – 239.

Lublinsky, B. (2001). Achieving the ultimate EAI implementation. Part 2:

Message-Level Integration. EAI Journal, 3, 18 – 31.

Mann, J. E., (2000). Workflow and enterprise application integration (EAI). EAI Journal, 2, 89 – 103.

Ren, F. (2002). The market place of enterprise application integration (EAI). EAI Journal, 4, 27 – 39

Abstract:

EAI is the process of integrating multiple, independently developed applications that may use incompatible technology and need to remain independently managed. Implementing EAI requires that information be moved between applications. This includes setting the applications up to send, receive, and react to information. Too often, companies waste a great amount of time, money, and other resources to begin exploring the potential benefits of EAI. This paper will provide a detailed review that demonstrates

the advantages of implementing EAI, the challenges that can be met, and the answers to those challenges.

Introduction:

EAI originated with the installation of enterprise resource planning (ERP) systems on a wide scale in the early 1990s. Companies had existing applications they wanted to leverage in the context of the ERP applications, and they could only do this by introducing EAI. Therefore, EAI has been very much a user-driven market.

Why companies are implementing EAI?

The trend toward EAI is a logical progression. Companies once used client/server technology to build departmental applications, but later realized the gains in linking multiple business processes. Companies built distributed computing environments, only to find a competitive advantage in expanding those applications to include external business partners. Kridva wrote, “The biggest draw to EAI came from the increasing number of Web-related projects and the realization that the implementation time was decreasing. It is often more expedient to develop an EAI strategy that reuses applications rather than starting from scratch” (Kridva, 1999, p. 226).

EAI software has the ability to integrate applications within the enterprise as well as across enterprises. With the collaboration and e-commerce becoming key, EAI software cannot be confined by enterprise boundaries. EAI software supports e-commerce, traditional electronic data interchange (EDI), and other means of inter-inter-enterprise application integration (Kridva, 1999, p. 226).

The IT landscape of the typical enterprise is a diverse collection of infrastructure components, including various classes of middleware from multiple vendors, several classes of databases from many vendors, and a myriad of file types. Even with standardization efforts, the typical IT infrastructure remains diverse. EAI software adapts to the changing IT infrastructure of the enterprise (Kridva, 1999, p. 226).

No company needs to increase its point-to-point interfaces. However, no one topology meets all the needs. The hub topology is efficient, but the bus, pipeline, network and other topologies also meet integration challenges. EAI software supports any number of topologies and does not limit the enterprise to a single, compromising hub topology (Kridva, 1999, p. 227).

Application integration is complex. EAI software handles this complexity without requiring users to exit out to manually coded libraries and routines. The elimination of programming even in the face of daunting complexity, characterizes EAI software (Kridva, 1999, p. 228).

How is EAI segmented?

EAI software is segmented in five hierarchical categories of integration:

Business process integration provides the highest level of abstraction and adaptability for an EAI solution. These solutions enable business managers to define, monitor, and change business processes through a graphical modeling interface. Business process integration requires all of the underlying integration services. Some vendors provide only the process integration, and their products sit on top of other EAI frameworks; others provide the complete solution.

Application integration provides a framework for a collection of technologies that together provide near real-time integration. The framework includes underlying platform integration technology; event integration through message brokers that provide data translation; transformation and rules-based routing; and application interface integration provided through application adapters to SAP and other ERP packages. Integration framework provide added value by abstracting the complexity of creating, managing, and changing the integration solution.

Component integration enables new functionality to be easily combined with ERP packages and existing client/server and legacy applications. Application servers provide this capability. Application servers also generally provide common services for security, transactions, load balancing and failover, connection pooling, state and session management, and data access to relational and non-relational sources. These services are also necessary for integration solutions.

Data integration products generally fall into two categories. The first includes database gateways that provide that provide SQL access to heterogeneous data sources. These are synchronous data access products and require that application developers have knowledge of the underlying database schemas. The second category of products provides tools for extracting, transforming, moving, and loading data. Data integration tools extract and load data directly, bypassing application logic.

Platform integration provides connectivity among heterogeneous hardware and operating systems, and technologies that provide platform integration include messaging. Most integration solutions require both asynchronous and synchronous connectivity (Ren, 2002, p. 29).

Challenges in implementing EAI

Nearly all of e-economy companies are running on the data in today’s integration systems, which means that they will require some form of integration. This is why EAI software implementation is so popular. Yet when implementing EAI, the problem lurks that various integration issues will arise at some point as an integral part of each project.

Pre-implementation approaches deal with the product selection based on developed sense of IT requirements. The biggest strategic issue is whether to create an overall EAI architecture into which future application integration will fit, or to build a flexible infrastructure into which integration technology can be added on an as-needed basis. Rodriguez warned, “Some IT managers that argue for comprehensive architecture say it provides the only method to address the business logic of the system, which guarantees efficient and complete integration. Others lean toward a flexible, modular architecture saying that a complete overhaul of the architecture is costly and unnecessary since their method achieves the same integration in less time. Selecting the right EAI software can be very risky in such a rapidly changing marketplace” (Rodriguez, 2002).

It is important to note that implementing EAI requires that IT people learn about the new software system. The learning process associated with implementing EAI makes the initial integration much more challenging than previously imagined. Most users must be trained so they can acquire all the skills needed to properly use the different features of the new EAI software. Yet, it may take more than the projected time for the learning process to complete. Meanwhile, companies struggle to restore employees’ confidence and minimize downtime (Lee & Bass, 2001, p. 68 – 69).

Another concern in EAI implementation is the customization required to effectively execute EAI software. Nearly all EAI implementations require some custom-developed code to automate the business processes that drive the applications. The majority of integration projects include legacy systems, for which no predeveloped connectors are available. Where the applications are homegrown, the challenges are higher still. According to Rodriguez, “There is an emerging recognition that EAI is not as simple brochureware makes it sound,” Rodriguez wrote. “truthfully it’s more complicated than that. The fact that vendors have formats laid out doesn’t mean they have all the integration and processes worked out” (Rodriguez, 2002).

When evaluating integration approaches, producer’s and customer’s integration may be a tough dilemma. Because producers are responsible to create messages when business events occur, the governing principle for producer’s integration is minimizing code changes. The real challenge, when the event is propagated from the database, a database trigger can send the message associated with that database change. According to Mann, “The difficulty associated with this approach can force companies to modify code of the producer application to send messages. The setback related to consumer’s integration is to preserve existing data validation and manipulation code” (Mann, 2001, p. 95). Consumer’s integration occurs via adapters, which are layer between the messaging middleware and application, any misconfiguration can disrupt the mapping between the integrated application and the messaging middleware native interfaces.

EAI implementation challenges organizations with costs on a scale that they have never attempted before and requires a corresponding large infrastructure investment to ensure integrity of mission-critical transaction processes. EAI costs come in three components: architecture, integrations, and operations. Architecture costs are expenses related to the initial deployment such as integration development, execution and operations environments. Integration costs are related to the development of interfaces and collaborations between systems. Operating costs include on-going operations and maintenance of the EAI system for architecture and integrations. Operating costs generally are driven by the number of interfaces that need to be maintained and rise as more interfaces are put into production (Mann, 2001, p. 95).

Solutions to challenges

Despite all the pitfalls of implementing EAI, IT managers must find ways to ensure their EAI software can successfully work to increase development productivity, to allow changes to the business and technical landscape to occur with minimal rework and impact on production systems, to shorten time for development of future resources, to provide improved access to timelier and more accurate data across a distributed environment while minimizing redundancy, finally to promote the management of integration at the business process level and allows for real-time and historical analysis of business conditions and performance. In order to be successful, IT managers have to find a solution to each specific problem that they can encounter.

According to Kridva, “The best solution to integration approaches is carefully consider the advice of well-meaning experts” (Kridva, 1999, p. 224). Also, organizations should ask themselves questions specific to their situation: What business process are we automating? What are the work flows, the business processes and the systems affected? Before embarking on EAI implementation, organizations must know every detail about their business process.

It does take some time for individuals to learn the new features of any new software system. It is not any different with EAI. Organizations must clearly arrange training sessions for employees to master the techniques of EAI software. Kridva wrote, “ When companies go the extra mile to educate their employees about EAI, they will later enjoy greater performance in project development and assimilation ” (Kridva, 1999, p. 226).

The need for customization may create a numerous challenges since few IT staffers have EAI knowledge available in-house. “In recognition of the ongoing need for customization, some companies are beginning to assign responsibility for these issues to an ‘enterprise architect’. These companies recognize that you can’t solve the problem just one, and assigning someone the responsibility for a strategic development can be worthwhile” wrote Kridva (Kridva, 1999, p. 227).

Effective data analysis and data management is a necessity if EAI projects are to be successful dealing with producer’s and consumer’s integration issues. Korzeniowski wrote, “Data analysis is essential to support a truly integrated supply chain. Where systems cannot be integrated with others in the supply chain because data is unstructured or poor quality or stored under codes alien to other businesses, then the supply chain links break and, ultimately, customer service suffers” (Korzeniowski, 2001, p. 120).

Customization and consulting can escalate the cost of implementing EAI strategies to a level that some organizations are not willing to go. According to Korzeniowski, “One way to skirt this problem is to provide cost and ease-of-use incentives for each division so they can easily buy products from their desktops at corporatewide volume pricing,” he also added, “The bottom line is that EAI is a strategic investment that can reduce costs, not only over the long run, but often on the very first project” (Korzeniowski, 2001, p. 121).

Future Trends

The driving force behind the burgeoning of EAI software is not only the possibility processes integration, but also a confluence of business and technology. A desire to extend the reach of business systems – from inside a corporation to its business partners and beyond – is driving the EAI market into three specific future trends.

One is to add process integration and monitoring capabilities to EAI systems. EAI vendors who are focused on the infrastructure part of adapters and translators are quickly moving to provide that capability.

The next one is consolidation and partnering. Several EAI vendors will work together to collaboratively sell and implement EAI applications of packaged integration applications to customers all over the world.

Finally, new vendors are coming into the space. Systems integrators who built tools are trying their solutions into the EAI market. The data integration and data warehouse players are moving in. The application server vendors are adding integration capabilities. Soon, everyone will want to be in the EAI space because it speaks to a business problem. Everyone has a piece of technology that can solve it (Lee & Bass, 2001, p. 68).

Conclusion

Most of today’s large companies run hundreds of different computing applications that were not designed to talk to other systems. So do their suppliers and buyers. EAI applications – even though they bring a great deal of challenges during the implementation phase – close the gap by pulling information from applications. Such information can be stored on rigid ERP systems – and sending it to a server that “brokers” the data as a message that can be understood by the receiver.

References

Girard, K. (200) Enterprise integration software’s new home on the Net. EAI Journal, 2, 111 – 120.