Divide and Conquer - Organizing Component-based Adaptation

Abstract

This paper introduces a divide and conquer approach for organizing
the adaptation of distributed applications in a potentially large number of
interacting middleware instances. In such an environment, a centralistic
and static adaptation reasoning i) is inadequate and ii)
gives the same priority to all applications. The divide and conquer method
aims at minimizing the interference between running applications, allowing
users to weight the priority of
applications, and organizing the
adaptation and the reasoning about the adaptation in a decentralized and
flexible way.