Enterprise Application Integration Controller

Enterprise Application Integration Controller Dr.Irfan Hyder

Syed Moinuddin Balkhi

Karachi, Pakistan. [email protected]

Karachi, Pakistan. [email protected]

Abstract— For business processes to span across organizational boundaries, ERP systems should integrate with multiple applications such as legacy systems, Electronic Fund Transfer Systems (EFT), and the Internet. Middleware is been promoted as a communication gateway those ingrates disparate systems transparently. Each ERP vendor is marketing it’s own middleware. Middleware is principally being used for business process integration. Middleware is additionally being used internally by organizations for integration with legacy applications and Internet portals. At present, these off-the-shelf middleware’s do not ensure recovery from message loss in instances when there is a message server failure, network failure, hardware failure, database failure or an incomplete message transmission. Message inconsistencies generally trigger issues such as missing number ranges for different objects e.g. (sale order, invoices), data integrity problems and non-reconciled disparate business systems. ERP vendors partnered with hardware vendors, by propagating clustered hardware solution for solving the message loss problem. However, this solution does not guarantee recovery from message loss; in addition it requires additional hardware and is an expensive proposition. This also necessitates that a homogeneous operating system environment be maintained. Also, it has been determined through experience that fail over on application server instances is not recommended for performance reasons. These instances need more resources for the transparent switchover process. . Since middleware is in the infancy phase, at present it does not have the capability to recover from message loss in case of non-availability of any of the systems. The motivation of this research is to find a software solution to minimize message loss when disparate systems are communicating with each other using a middleware. In order to demonstrate minimization of message loss, an environment where SAP is being used as an ERP, and integration with SAP is being achieved through middleware, a JAVA swing based application is proposed that will ensure minimum message loss, and would function like an EAI controller. The core idea behind controlling and monitoring the message communication from a remote location-using controller is to provide a robust, low cost and high

availability environment to EAI. The controller will monitor all the messages communication between User-End and SAP R/3 and the re-transportation of message to middleware from controller’s repository in case of any message lost or complete middleware failure. The results show the presence of such EAI controller significantly reduces message loss in the event of nonavailability of any of the systems. Message recovery and message reapplication can be performed transparently. In addition, the research shows JAVA RMI is ideally suited for situations in which separate parts of a single program can exist in multiple Java environments on multiple machines. To the client and server, the program operates much like it is local to their environment and within a single memory space. The future goal of the research is to develop strategies for making the controller a robust application capable of handling diverse platforms and user interface that automatically generates java code based on UML modeling.

INTRODUCTION (HEADING 1) Enterprise Resource Planning systems (ERP) are commercial software packages designed to assist organizations to integrate existing administrative systems in the areas of finance, human resources, supply chain information and customer information. Since the late 1990s, there has been an increasing use of ERP in most large businesses and organizations and more recently, in the Oil & Gas Sector in Pakistan. The widespread use of ERP is a response to the needs of businesses and organizations to replace legacy (batch mode systems) and achieve integration of different organizational functions and systems. I.

This research aims at minimizing message loss, specifically when a middleware is being used for integrating Enterprise systems. It proposes a solution to minimize message loss, thereby ensuring high availability and data integrity of the transactions. Integration of business process across organizational boundaries will result in tighter coupling of supply chain activities. In this chapter, the endeavor is to outline the details and the boundaries of the research. It provides a background to

the research and also an overview of the research undertaken previously. The justifications of the research, its significance and its limitations are also discussed. In recent years, in order to retain customers in a competitive environment and to effectively manage supply chain activities, companies are providing additional facilities to customers and suppliers, which include e-commerce, b2c/b2b connectivity and value-added services. These factors, combined with regulators’ interventionist strategies to adopt modern ways of doing business, have driven the need for companies to improve the administrative efficiency of their operations. ERPs have been introduced into corporate sector with the aim of improving and integrating administrative systems efficiency while at the same time, providing focus on improved customer service by offering ecommerce strategies.

communication and conversion purposes. Message loss due to middleware failure, network failure, hardware failure, and database failure or due to an incomplete message eventually can affect the customer-end results. Inconsistency is introduced in business data, number ranges maintenance for different objects (sale order, invoices) and finally, the integrity between business processes. Specifically, this research aims to develop a methodology for minimizing message loss when integrating ERP systems with disparate systems using middleware. In addition, it examines literature on ERP implementations in corporate context with a consideration to review integration issues and explore existing tools, mechanisms available for integration of ERP systems with disparate systems more specially by using middleware. A literature review was undertaken to identify:

ERP vendors have started focusing on releasing products that support the integration with other applications and ERP systems. SAP® has released Netweaver that supports integration services. E-commerce is being offered to customers in Pakistan. With the availability of Internet services to majority of corporate customers, the corporate sector is offering value-added services through the Internet. Majority of the companies have implemented ERP systems in manufacturing, utilities and the oil and gas sector in Pakistan. These have been technically successful implementations, i.e. in working as standalone systems. In order to reap business benefits, companies are planning and implementing integration across organizational boundaries.

·

The usage and the meaning of key terms and concepts with regard to managing integration of ERP system with external systems and to establish the possible range of issues associated with implementing an integration solution.

·

The contemporary research associated with integrating ERP systems with external system in relation to more specifically SAP system using middleware.

·

Solving an EAI problem that requires communication with diverse applications on diverse platforms

As noted above, ERP systems have been introduced into industry with the aim of improving efficiency of business operations, with added benefits of improved customer services. To allow its ERP applications to create a business process that spans multiple applications, middleware has been introduced that offers two primary things: a way to specify and implement the logic driving that business process, and some mechanism for communicating between the ERP applications the business process uses. Different[56] adapters provide different communication mechanisms, so a message might be acquired by accessing a Web service, reading from a file, or in some other way. However [56], automation brings in new challenges of creating automated business processes that rely on diverse systems. Core messaging becomes complicated with complex business scenarios. Monitoring [56] message exchanges triggered by business-oriented information about a running process and business activity services to work with trading partners becomes increasing tedious. The message then processed may be lost leading to transactional loss and data integrity issues.

·

The contemporary research is focused on system-tosystem communication, supporting business processes that depend on integrating diverse software.

II. A.

SCOPE AND AIM OF THE STUDY

Template (Heading 2) Situated in the context of integrating ERP systems using middleware, which are used typically in complex multisystem, heterogeneous environment is used for

B. Existing Middleware The solutions proposed by software vendors are clustering based and are focusing on high availability of databases and backup of middleware to recover the lost transactions from databases manually. The solution proposed in the thesis is a low cost, platform independent solution. It ensures the high availability of messages at software application-layer instead of targeting hardware and databases layer. There is no need to ensure the high availability of individual adapters/connectors as the controller is monitoring and storing different kind of messages that pass through the middleware The proposed solution can also reduce the cost of maintaining database backups and clustering environments. Auto recovery of lost message without human intervention makes the solution more reliable when compared to commercially available solutions. C. Significance This research identifies the vulnerabilities of middleware, which commercial vendors tend to ignore or underscore. The research highlights how functionality of middleware can

be improved upon by introducing a new component called EAI Controller that will minimize message loss. Integration infrastructure complexity is growing day by day with the addition of new applications. In a typical integrating environment, each application talks with only one middleware and is not aware of other applications; a transition from point-to-point to hub-and-spoke architecture. Even so, many complex business transactions are processed today without message storing. In [56] a large network, the system might be maintaining many thousands of connections in a ready-to-use state. If one part of the system suffers a problem, many parts of the system become unusable. A abstraction layer in terms of a EAI controller is proposed that will act like a hub for all incoming and outgoing transactions and which will initiate transaction recovery by storing and forwarding message in case of unavailability of any of the systems. This research is limited to environments in which SAP is being used as an ERP and JAVA RMI is used to create the middleware, controller and application server. Only limited scenarios were formulated and tested. The application designed is just a prototype that needs considerable improvement in order to become a portable product. III. THE METHODOLOGY Most commercial middleware do not have an inherent mechanism to recover from message loss. If the message loss occurs during the transmission of messages between enterprise systems, then there is no retransmission mechanism. Message loss can eventually affect the customer results (client), business data, number ranges maintenance for different objects (sale order, invoices) and finally, the integrity between business processes. A methodology is proposed which will help to minimize message loss in instances when multiple systems are communicating with each other using the middleware. For reliable communication, messages lost must be detected and re-transported to Enterprise systems. The proposed methodology is based on following four functions:

Figure-1 Methodology Overview

A. Message Monitoring A middleware between clients and SAP R/3 is required, to translate different types of protocols and messages communicated to SAP R/3 in a form that is acceptable for SAP R/3. Middleware acts as a hub between SAP R/3 and other Non-SAP products for all types of communications with SAP R/3. However ensuring reliable communication in middleware remains a challenging problem. Middleware can become a single point of failure (SPOF) for all types of communications. Failures can be raised due to a problem in middleware, network failure or unavailability of SAP R/3. In any case, there is need for a mechanism for lost message detections and recovery. The basic theme of the solution is to detect lost messages and there is a separate well-built monitoring module that can closely monitor all the communication between clients and R/3. For message monitoring, a software module called EAI Controller is proposed. Controller is a lightweight module developed in Java, and which is located in a separate dedicated machine. Controller continuously monitors the middleware in order to be equivalent with the current status of the Middleware. Message monitoring performed remotely by the controller is based on calls to following java methods located in connector class (middleware) using RMI. ·

notification()

·

Message Monitoring

·

readMessage()

·

Cache Mechanism

·

check_Result()

·

Cache Load Balancing

·

Error Detection and Recovery

Whereas the algorithm and the implementation details are explained in the following sections: ·

Algorithm

·

Implementation Details

The call to method notification() returns a pointer having a value 0 or 1. As soon as any message arrived in middleware from a client, class connector sets the global variable PTR of Type Integer to 1. Controller reads variable PTR using method notification () for detecting that a message has arrived and further processing is required at the controller end to make message transmission guaranteed. Variable PTR is initialized back to 0 after transmitting the message safely to R/3. In second step, after detecting that a message is sent by the client to the middleware, controller triggers the method

readMessage() located in connector class (middleware). Method readMessage() reads the message sent by client and stores it in an array with a message arrival timestamp in milliseconds and this information is returned back to controller using method readMessage(). In third step, method checkResult() is called by the controller to monitor the results received from the middleware from SAP R/3. Method checkResult() searches the array used for storing the results received from SAP R/3 in the middleware. In case of any message failure, the array will return no results with a pointer containing NULL value and controller then performs message recovery procedure. B. Cache Mechanism Controller stores all messages at a remote location using a cache called SAP message repository. For simplicity and simulation purposes all communications are stored in an access sheet using JDBC driver. This can be replaced by any commercial DBMS in business scenarios. All messages read by the Controller are stored in the message repository with the time stamp of message arrival in the middleware. The purpose of storing message at remote location is providing a repository for recovering all communications that can be recovered even if the middleware is unavailable or cache residing in the middleware gets inconsistent. C. Cache Repository To make the solution cost effective and faster, stored messages in repository get deleted by controller after getting the status that the message has been transported successfully using checkResult() method. D. Error Detection and Recovery The method checkResult() is a critical method to check the result of message that was sent to R/3. Any client that communicates with R/3, calls a method sendMessage() in middleware (Connector Class) to send the message to R/3. Method sendMessage() uses the Jco (Java Connector) to develop a link with SAP R/3, which in turn calls a BAPI ZFLIGHT , developed in SAP R/3 using ABAP (Advance Business Application and Processing).

In case of successfully transporting the message to SAP R/3, controller gets the indicator of 1 or true from the middleware. Then the controller will delete the message from repository. This implies there will be minimum message storage in the repository and the repository will have only messages with error indicators, thus making the search for any particular message faster. Messages can be lost in following three main scenarios: ·

Loss of message due to network failure.

·

Message loss due to R/3 failure.

·

Message loss due to complete failure of middleware.

In first two scenarios message loss can be recovered by simply re-transporting the message again from controller to middleware, which in turn gets transported to SAP R/3. In the third scenario, the main communication mechanism itself will be unavailable; no further communication from client side is possible. All messages transported till the point of failure, if not completed successfully, will be stored in controller message repository. Controller will wait for the middleware to become active again, and will re-transport all the messages again to the middleware in the same order in which they were received with the help of time stamp IV.

ALGORITHM

1.

Client sends message to middleware.

2.

Controller checks message repository and send all lost messages found in repository to middleware one by one.

3.

Controller checks status of messages send by client in middleware

4.

If message found goes to Step 5 else go to Step 3.

5.

Controller saves repository.

message

in

SAP

message

6.

Middleware sends message to R/3.

Results of message are returned by R/3 in middleware, using a BAPI, the reciprocal process is done by method sendMessage (), which returns array of method, the array is globally defined class and can also be read by the controller to monitor the results.

7.

R/3 Sends back the results of query back to middleware.

8.

Controller checks the result.

9.

If Results = OK go to step 10, else go to step 2.

The method checkResult() searches the array to detect the successful completion of message process. In case of any failure or an empty array, method Result () will return an indicator of 0 or null to controller with the time stamp. Controller searches for particular time stamp id in message repository and recalls the sendMessage() to send the same message again to R/3.

10. Delete the message from repository transferred successfully. V. IMPLEMENTATION DETAILS The proposed software solution is using JAVA SWING technology for developing the graphical user interfaces and RMI for communication across the distributed system. Client

will be able to communicate with SAP from any remote location using a middleware server, developed in Java. It is implementing SAP Java Connector for communication between middleware and SAP R/3. Middleware server used here is to facilitate communication between client, SAP R/3 and Controller and is accessible from any location i.e. from a web portal or a simple remote desktop machine. Controller and middleware server are low cost solutions and are able to operate from any desktop machine. The basic purpose of controller used here is to monitor the all messages communication between client and SAP R/3 and the re-transportation of message to middleware from controller’s repository in case of any message failure. Data requested by user (client) is received by middleware and also transported to controller, which stores it in a central message repository. Messages from middleware are transported to SAP R/3. Controller re-confirms the SAP for a successful data receiving by monitoring the results of attempt to communicate with R/3. In case of any failure, data can be transported back to middleware from controller to retransport it to SAP R/3. Data requested by user is transmitted to middleware from SAP, and a message of a successful arrival of data from SAP is also sent to controller. Controller keeps storing all messages conversation in central message repository until the data safely transported to user end.

Controller communicates with middleware using RMI to verify the successful arrival of messages in SAP R/3. In case of any failure to send message in SAP R/3 from middleware, message can be re-transported to middleware server from controller, and finally that can be transported to SAP R/3 again. Java Connector implemented in middleware is using the BAPI and RFC to communicate with SAP R/3. A Remote Function Call (RFC) is the call of function module in a partner system. The caller is the RFC client and the called partner is the RFC server. RFC is based on the known RPC model from the UNIX-TCP/IP environment. RFC in SAP environment is based on a CPI-C interface implemented by SAP. SAP BAPIs provide developers with stable, reliable interfaces for accessing the functionality offered by SAP business objects. BAPIs can reduce the development cost by providing built-in functionality through predefined methods of the SAP business objects. These methods can reduce programming overheads in deployment phase. The Business Object Repository (BOR) contains the SAP business object types and SAP interface types as well as their components, such as methods, attributes, and events.

The whole model is based on the following components: ·

Client To send and receive messages, to and from SAP R/3 via middleware.

·

Middleware A Server that implements the SAP Java Connector, listen requests from client, transfer it to SAP R/3 and vice versa.

·

Interface provides methods definitions for both client and middleware server.

·

Implementer implements all the methods provided by Interface class.

·

Controller communicates with middleware, to monitor all messages between client, middleware and SAP R/3 with storage capability of storing messages in a central message repository with a unique number, date and time stamp.

Client, controller and middleware server communicate with each other using RMI calls from remote locations. Controller is responsible to store each message transferred to SAP R/3 in a central message repository (CMR) using JDBC with a unique identification number and time stamp.

Figure-2 System Diagram

VI.

PERFORMANCE COMPARISON

Following is the percentage factor breakdown before implementing proposed solution.

Components Middleware

Failure Criticality 100%

Failure Chances 100%

R/3

100%

100%

Client

80%

0%

Network

80%

100%

Total

360

Table 1: Failure Criticality Before Implementing Controller

Conclusion— In this research, only recovery of message loss is studied in scenarios where SAP R/3 is been used as an ERP, and middleware is used to communicate between client and ERP. The future goal of the research is to develop strategies for making the controller a robust off-the-shelf application capable of handling diverse platforms and user interface that automatically generates java code based on UML modeling. REFERENCES

Components

Failure Criticality

Failure Chances

Middleware

100%

100%

R/3

100%

100%

Client

80%

0%

Network

80%

100%

Total

[1]

[2]

[3]

360

Table 2: Failure Criticality After implementing Controller

[4]

[5]

100 [6]

80 [7]

60

Before

40

After

[8]

20 0 Middleware

[9]

Controller

Network [10]

Figure –3 Failure Ratios [11]

VII. CONCLUSION & FUTURE WORK This research proposes message-controlling mechanism, which would minimize message loss between middleware and ERP systems and also, augments recovery from messages loss in case any of the system fails or becomes temporarily unavailable. The results show that the proposed mechanism minimizes message loss by selectively storing unprocessed messages and reprocessing the message. Furthermore, the result shows that in proposed mechanism, the recovery from message loss is 99 %, which is greater than any of the hardware clustering mechanisms available.

[12]

[13]

[14]

[15]

Daneva M,Using ,“Maturity Assessments to Understand the ERP Requirements Engineering Process,” from proceeding off IEEE Joint International Conference on Requirements Engineering, September 12th 2002. Jen-Her Wu, Shin-Shing Shin, Chi-Cheng Wu, "COTS-Based Systems: A Methodology for Evaluating Data and Output Misfits," from proceedings of the 38th Annual Hawaii International Conference on System Sciences, January 12th 2005. L. Brehm, A. Heinzl, M. Markus, "Tailoring ERP Systems: A Spectrum of Choices and their Implications,” from proceedings of 34th Annual Hawaii International Conference on System Sciences (HICSS-34)-Volume 8, -2001. Daneva M., “Lessons Learnt from Five Years of Experience in ERP Requirements Engineering,” from proceedings 11th IEEE International Requirements Engineering Conference- 2003. Gulla J A, “Understanding Requirements in Enterprise Systems Projects,” from proceedings of Requirements Engineering Conference,” 12th IEEE International 2004 . Daneva M, “Establishing Reuse Measurement Practices in SAP Requirements Engineering,” from proceedings of Fourth International Conference on Requirements Engineering – 2000. Daneva M, Wieringa R, “Requirements Engineering for Crossorganizational ERP Implementation: Undocumented Assumptions and Potential Mismatches,” from proceedings of 13th IEEE International Conference on Requirements Engineering – 2005. Rolland C, Prakash N, “Matching ERP System Functionality to Customer Requirements,” from Proceedings of Fifth IEEE International Symposium on Requirements Engineering – 2001. Gulla J A, Brasethvik T, ”On the Challenges of Business Modeling in Large-Scale Reengineering Projects,” from proceedings of Fourth International Conference on Requirements Engineering – 2000. Shanks G, Parr A N, Shanks G , “A Taxonomy of ERP Implementation Approaches,” from proceedings of 33rd Hawaii International Conference on System Sciences-Volume 7, January 2000. Etien A, Salinesi C, “Managing Requirements in a Co-evolution Context,” from Proceedings of 13th IEEE International Conference on Requirements Engineering –2005. Gibson N, Christopher P. Holland, Light B, “Enterprise Resource Planning: A Business Approach to Systems Development,” from proceedings of Thirty-second Annual Hawaii International Conference on System Sciences, Volume 7 –1999. Puschmann T, Alt R, “Enterprise Application Integration, The Case of the Robert Bosch Group,” from proceedings of 34th Annual Hawaii International Conference on System Sciences (HICSS-34)Volume 9 – 2001. Huber T, Alt R, Osterle H, ”Tailoring ERP Systems: A Spectrum of Choices and their Implications,” Proceedings of 33rd Hawaii International Conference on System Sciences-Volume 7 -2000 . Themistocleous M, Irani Z,Peter E.D, “Extending the Information System Lifecycle through Enterprise Application Integration: A Case Study Experience,” from proceedings of the 37th Annual Hawaii International Conference on System Sciences (HICSS'04) - Track 8 2004

[16] Themistocleous M, Irani, O'Keefe, R Paul , “ ERP Problems and Application Integration Issues: An Empirical Survey,” from proceedings of 34th Annual Hawaii International Conference on System Sciences ( HICSS-34)-Volume 9 -2001 [17] Riet R V, Janssen W, Gruijter P D, “Moving from Database Systems to ERP Systems,” from proceedings of 9th International Workshop on Database and Expert Systems Applications (DEXA'98) [18] Sedigh-Ali, Ghafoor A, Paul S, “Metrics-Guided Quality Management for Component-Based Software Systems,” from proceedings of 25th Annual International Computer Software and Applications Conference, Publication –2001. [19] Kujala , Kauppinen M, Lehtola L, Kojo T, ”The Role of User Involvement in Requirements Quality and Project Success,” from proceedings 13th IEEE International Conference on Requirements Engineering -2005. [20] Joseph Weiss, Sue Newell, Jacky Swan, "Minitrack Introduction," hicss, p. 190, from proceedings of the 39th Annual Hawaii International Conference on System Sciences (HICSS'06) Track 8, 2006. [21] Toni M. Somers, Klara Nelson “The Impact of Critical Success Factors across the Stages of Enterprise Resource Planning Implementations” Proceedings of the 34th Hawaii International Conference on System Sciences - September 2001. [22] Paul Hawking, Andrew Stein,“Revisiting ERP Systems: Benefit Realization,” from proceedings of the 37th Hawaii International Conference on System Sciences – 2004. [23] William Wagner, Yvonne Lederer Antonucci, “An Analysis of the Imagine PA Public Sector ERP Project,” from proceedings of the 37th Hawaii International Conference on System Sciences –2004. [24] She-I Chang ,“ERP Life Cycle Implementation, Management and Support: Implications for Practice and Research,” from proceedings of the 37th Hawaii International Conference on System Sciences – 2004. [25] Marinos Themistocleous, Zahir Irani, Robert M. O’Keefe, Ray Paul, ”ERP Problems and Application Integration Issues: An Empirical Survey,” from proceedings of the 34th Hawaii International Conference on System Sciences - 2001. [26] Nicola Gibson, Christopher P. Holland and Ben Light, “Enterprise Resource Planning: A Business Approach to Systems Development,” from proceedings of the 32nd Hawaii International Conference on System Sciences – 1999. [27] Lars Brehm, Armin Heinzl, M. Lynne Markus,“Tailoring ERP Systems: A Spectrum of Choices and their Implications,” from proceedings of the 34th Hawaii International Conference on System Sciences - 2001. [28] Jen-Her Wu, Shin-Shing Shin and Chi-Cheng Wu,“COTS-based Systems: A Methodology for Evaluating Data and Output Misfits,” from proceedings of the 38th Hawaii International Conference on System Sciences – 2005. [29] Marie-Claude Boudreau, “Learning to Use ERP Technology: a Causal Model,” from proceedings of the 36th Hawaii International Conference on System Sciences 2003 S’03) [30] Guy Gable, Taizan Chan,“An ERP Maintenance Model,” from proceedings of the 36th Hawaii International Conference on System Sciences 2003 SS’03). [31] Sanna Laukkanen, Sami Sarpola, Petri Hallikainen, “ERP System Adoption - Does the Size Matter,” from proceedings of the 38th Hawaii International Conference on System Sciences -2004. [32] Lester A., “Applications Integration: Is it Always Desirable,” from proceedings of the 37th Hawaii International Conference on System Sciences – 2004. [33] Thomas F, L. Goodhue, “Understanding the Plant Level Costs and Benefits of ERP: Will the Ugly Duckling Always Turn Into a Swan,” from proceedings of the 33rd Hawaii International Conference on System Sciences – 2000. [34] Christian Janiesch, Alexander Dreiling, Ulrike Greiner, Sonia Lippe, “Integrated Configuration of Enterprise Systems for Interoperability –

[35]

[36]

[37]

[38]

[39] [40]

[41] [42] [43] [44] [45] [46]

[47] [48] [49] [50] [51] [52] [53] [54] [55]

[56] [57] [58]

Towards Process Model and Business Document Specification Alignment,” from proceedings of the 10th IEEE International Enterprise Distributed Object Computing Conference (EDOC'06). Marinos Themistocleous, Zahir Irani, “Evaluating and Adopting Application Integration: The Case of a Multinational Petroleum Company,” from proceedings of the 35th Hawaii International Conference on System Sciences –2002. A. Botta, B. Lazzerini, F. Marcelloni, “Integrating service-oriented technologies to support business processes,” from proceedings of the 2005 Seventh IEEE International Conference on E-Commerce Technology Workshops (CECW’05). Francesco di Cugno, Tommaso Di Noia,Eugenio Di Sciascio, Azzurra Ragone Politecnico di Bari, ” Concept Covering for Automated Building Blocks Selection based on Business Processes Semantics,” from proceedings of the 8th IEEE International Conference on E-Commerce Technology and the 3rd IEEE International Conference on Enterprise Computing, E-Commerce, and E-Services -2003. Jinghai Rao, Dimitar Dimitrov, Paul Hofmann, Norman Sadeh , “A Mixed Initiative Approach to Semantic Web Service Discovery and Composition: SAP’s Guided Procedures Framework,” from proceedings of IEEE International Conference on Web Services (ICWS'06). Unknown Author,” ASAP Project Management Framework v2.2.doc,”not published. Maya Daneva, Roel Wieringa,” A Coordination Complexity Model to Support Requirements Engineering for Cross-organizational ERP,” from proceedings 14th IEEE International Requirements Engineering Conference (RE'06). Unknown Author,” ALE Introduction and Administration Release 4.6c SAP®,” not published. William Adams ,”Scalability Options of SAP NetWeaver,” not published. Dan Woods, “Jeffery Word SAP® Netweaver for Dummies,” not published. Jeff Gebo, “SAP An A-to-Z Guide to SAP NetWeaver Developer Studio,” not published. Unknown Author, “BMC Control M Empowering the Net weaver Solution,” not published. Unknown Author,” High Performance Sap Netweaver Business Intelligence Solutions Based on IBM DB2 9 and IBM system Solutions for BI accelerator from SAP®,” not published. Unknown Author, “END-TO-END PROCESS INTEGRATION WITH SAP NetWeaver™,” not published. Unknown Author,”SAP NetWeaver™,” not published Unknown Author, “Aris for SAP Netweaver The Business process design Solution for SAP Netweaver,” not published. Unknown Author, “TABC43 Data Transfer SAP®,” not published. Unknown Author, ”Remote Method Invocation (RMI),” not published. Unknown Author, ”Sun Development Network Java Remote Method Invocation - Distributed Computing for Java,” not published. Teresa Jones,”E2E Technologies Bridge v3.6,” not published. David Chappell, “Understanding BizTalk Server 2006,” not published. Lee Gavin, Gangadhara Rao Akana,, Axel Angeli, Sreedhar Reddy Bhumireddy, Peter Gersak,Nizam Ishmael,Hari Kumar, Nagumalla, Viral Ved ,”A B2B Solution using WebSphere Business Integration V4.1 and WebSphere Business Connection V1.1,” not published. Unknown Author,” BizTalk Server 2004 Technical Guide for High Availability,” not published Unknown Author,” SAP XI Release 3.0, D o c u m e n t V e r s i o n : 1 0 . 0 0,” not published. Lee Gavin, Gangadhara Rao, Axel Angeli, Reddy Bhumireddy, Peter Gersak, Nizam Ishmael, Hari Kumar, ”A B2B Solution using

WebSphere

Business

Integration,

V4.1

and

WebSphere

Business,Connection

V1.1,”

not

published.