A model and methodology for distributed integration

Abstract

The authors identify some of the problems in integration. They give a survey of the various integration strategies that have been proposed in the literature. The success of these integration strategies depends heavily on the characteristics of the application and choosing the appropriate strategy has been for the most part done subjectively. What is needed is a general model which allows the authors to quantitatively explore the tradeoffs and select a strategy or combination of strategies. They propose such a model for the integration process based on graph theory. Their model provides a quantitative method for comparing the different integration strategies to be used for integrating a particular system. Another issue in systems integration is maintaining the consistency of assumptions made by different software engineers. They suggest a technique to keep track of consistent assumptions during the development process.<>