A Component Platform for Experimenting with Autonomic Composition

In this paper, we propose a component-oriented framework that can support autonomic computing and in particular bio-inspired approaches. Starting from the Grid Component Model, a component model targeting at Grid computing and already featuring some autonomicity, we show how such a model can be used in a general autonomic computing context. Indeed the model provides hierarchical structure and reconfiguration for both functional and non-functional levels. This should ease the development of self-* and in particular, self-evolving applications. With our approach, even the autonomic strategies themselves can evolve. We consider this model and its implementation as powerful tools for easily experimenting autonomic behaviours.