Chapter 1   Overview


The iPlanet Application Server Enterprise Connector for R/3 is used for building and delivering scalable applications that integrate the application server with R/3 ERP (Enterprise Resource Planning) management systems.

This chapter introduces the iPlanet Application Server Enterprise Connector for R/3 and describes some fundamental concepts the administrator should be familiar with when setting up the connector for R/3. This chapter contains information about the following topics:

• Unified Integration Framework (UIF)

• Three-tier Application Model

• High-level System Architecture

• Enterprise Connector Tools for R/3

• National Language Support

Unified Integration Framework (UIF)

---

The UIF is an application programming framework that provides a single Application Programming Interface (API) to access different back-end systems. A connector is developed for each back-end system to allow communication between the UIF API and the back-end system, as seen in Figure 1-1. The UIF API is the only API necessary to access the back-end system.

Figure 1-1    Unified Integration Framework

The UIF enables development of server extensions that integrate with legacy systems, client-server applications, and third-party Internet solutions. These extensions provide a consistent access layer to disparate back-end systems, dramatically reducing development effort. The framework provides support for features such as object-pooling, and distributed state and session management.

A generic data repository is also part of the UIF, which is used to hold metadata parameters and other information about the back-end system. For example, the metadata often describes the physical connection between systems, the data that is available, and methods you can use to process data.

The implementation of each connector is specific to its the back-end Enterprise Information Systems (EIS) since each EIS is different.

UIF Services

The UIF is a component of the iPlanet Application Server. The iPlanet Application Server plays a prominent role in a three-tier application model. See the Three-tier Application Model for a description. The UIF mediates between the iPlanet Application Server application and the EIS (back-end) tier, namely the data sources and databases.

The UIF provides an API to access the following services:

• Runtime

• Data Object Services

• Repository and Metadata Services (not available to users)

Runtime

The UIF runtime services supply core services for resource management, thread management, communication and life cycle management, and exception management. The UIF runtime services understand and interpret metadata repository contents.

Data Object Services

The Data Object Services implements universal data representation common to all connectors. See the iPlanet Application Server Enterprise Connector for R/3 Developer's Guide for a description of data objects.

Repository and Metadata Services

The UIF repository and metadata services model a persistent information hierarchy that supports datatype definitions and inheritance. It also manages the instances and reuse of data objects from datatype definitions.

Three-tier Application Model

---

The machine and software involved are divided into the following three tiers:

• Client Tier

• Server Tier (iPlanet Application Server)

• EIS (Back-end) Tier

The connectors serve as an essential link allowing the server tier to communicate with the EIS tier, as shown in Figure 1-2. Communication between the application server and the back-end EIS is facilitated by the UIF API. This layer of functionality resides as an added layer to the iPlanet Application Server and enables data communication with diverse back-end EIS's in a seamless and uniform manner.

Figure 1-2    Three-tier Web-based Computer Model

Client Tier

The client tier is represented as the user interface. Requests for data originate here, represented by web browsers or rich clients (such as a Java applet).

Server Tier

The server tier is represented by an application server and optionally a web server such as the iPlanet Web Server Enterprise Edition. The server tier houses the business logic (Enterprise Java Beans and/or your application servlets), and provides scalability, high availability load balancing, and integration with a variety of data sources.

EIS (Back-end) Tier

The back-end tier is represented by an Enterprise Resource Planning (ERP) systems database or another back-end data system such as CICS.

High-level System Architecture

---

When the iPlanet Application Server startup process begins, the UIF loads the connector into the process. When the client requests data from the back-end system, a daemon process is started manually. The connector notifies the daemon of the request using a proprietary protocol. The daemon either creates a new worker process to process the request or passes on the request to an idle worker. The actual conversation to the back-end system is done by the worker process. The worker returns the results to the connector using the proprietary protocol. Figure 1-3 displays the system flow of the iPlanet Application Server.

Figure 1-3    System Flow

Enterprise Connector Tools for R/3

---

The Enterprise Connector Tools are as follows:

• Management Console - includes User Mapping and Data Mining Tools
  • User Mapping - allows you to map user IDs for access into the back-end system, and to edit and manage data sources.

    A data source represents the connection to the back-end system.

  • Data Mining Tool - includes capabilities such as determining the available functions in the back-end system, translating and reformatting data, and loading data into the data repository.

• Repository Browser - allows you to browse data in the repository. You can view the available functions (input and output parameters) for the back-end system.

The Management Console

The iPlanet Application Server Enterprise Connector for R/3 has an interactive Management Console Tool. The tool is used by the data source administrator to do the following:

• data mine function objects and data types from the back-end system metadata, and populate the repository with these data-mined definitions

• administer repository contents for creating new data sources, editing settings, and altering pooling settings

• populate user-mapping tables in the data repository

For more information on the Management Console, see The Management Console.

The Repository Browser

The Repository Browser allows the user to explore the repository contents. Repository contents are modified/populated using the import function. The import function is also available as a browser function and as a command-line tool, and imports XML under a specified repository node. Additionally, the export function is available as a browser function and as a command-line tool and exports a selected subtree as an XML document. For more information on the Repository Browser, see The Repository Browser.

---

Caution

The Repository Browser should not be used for editing even though import, export, and delete actions on repository nodes are enabled. Only advanced administrators should use the available functions.

---

National Language Support

---

The iPlanet Application Server Enterprise Connector for R/3 provides full support for building applications that use the prevalent and popular character sets such as ASCII, BIG5, SJIS and others. Chapter 3 describes how to modify a back-end data source for international use.