Abstract
The classic way of building a software today sim-plistically consists in connecting a piece of code calling a method with the piece of code implementing that method. We consider these piece of code (software systems) not calling anything, behaving in a non deterministic way and providing complex sets of services in different domains. In software engineering reusability is the holly grail, and specially the reusability of code from autonomus tools requires powerful compostion/integration mechanisms. These systems are developed by different developers and being modified inceremently. Integrating these autonomous tools generate various conflicts. To deal with these conflicts, current integration mechanisms defines specific set of rules to resolve these conflicts and accompalish integration. Indeed still there is a big chance that changes made by other developers, or they update their changes in order to make them compliant with other developers cancel the updates done by others. The approach presented here claims three contributions in the field of Hetrogeneous Software Integration. First, this approach eliminate the need of conflicts resolving mechanism. Secondly, it provides the mechanism to work in the presence of conflicts without resolving them. Finally, contribution is that the integration mechanism does not affect if either of the system evolves. We do this by introducing an intermediate virtual layer between two systems that introduce a delta models which consist of three parts; viability that share required elements, hiding that hide conflicting elements and aliasing that aliases same concepts in both systems.
Highlights
We present here a small scenario to show that how Virtual Heterogeneous Model Integration Layer (VHMIL) is created on a real time case study
The authors presented a conceptual co-evolution framework with help of running example for decrease in co-evolution effort and increase in co-evolution performance. It is a permanent effort in computer sciences, to find ways to integrate existing applications with new ones
This paper presents the current state of work in attempting to overcome the difficulties involved in application integration and reuse each other’s services
Summary
One way to integrate sub systems and share as much data between them is by resolving all of their conflicts. As the matter of fact, it works as a new central independent layer to facilitate sharing of features and data between both the systems. We already stated that it is a virtual layer, which implies that it does not affect the structure of both sub systems On the contrary, it takes domain specific language of two systems as input plus necessary requirements from their domain experts and defines a new composed language. VHMIL works as a filter of features between both the systems Domain experts from both systems identify the features that what should be and should not be shared. Domain experts have to decide which features need to be shared
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
