FRIEND Overview

Introduction

FRIEND (FRamework for the Integration of heterogeneous ENterprise Data) is an interdisciplinary project in the area of information systems.

The main objective of the FRIEND project is to enable System Wide Information Management (SWIM) by implementing a framework for the integration of various data managed by autonomous systems within a Community of Interest (COI).

User applications will be able to access local data in a secure and fully location-transparent way through the uniform interface of the Virtual Information Pool. The resulting system shall allow for queries spanning several user views and several local systems.

Current Situation - Physical Data Exchange

Information sharing among autonomous systems (which use different data models, different data exchange formats, different database management systems) is difficult without a consistent and system-wide concept of data integration.

Figure 1 shows the situation which can be found very often in large enterprises and in inter-enterprise communications nowadays: information sharing is done by defining extra interfaces for data exchanges. This approach is error-prone and costly.

Current situation

Figure 1: Current situation

FRIEND Solution - Logical Integration of Data

In FRIEND data are logically integrated (see Figure 2) thereby eliminating potential consistency related problems, i.e. no physical data exchange is required. Local data models are mapped to the corresponding component models (Double Arrow) which are then unified in the global data model.

User applications can rely on a standardised, object-oriented global data model of the Virtual Information Pool to access data spanning several local systems (Arrow).

Logical Data Integration

Figure 2: Logical Data Integration Using the FRIEND Integration Layer

Top