Using an Automated Planner to Control an Organic Middleware

Abstract

Organic Computing Systems feature self-organizing techniques to manage the rising complexity of distributed systems. This paper introduces an implementation of selfconfiguration, self-healing and self-optimization by means of an Organic Manager. The Organic Manager is based on an automated planner and integrated in a middleware. The Organic Manager unites formerly distributed self-x features to use synergetic effects. An evaluation with four scenarios compares the behavior of the system under two different planning models. We show that an automated planner can lead to an optimized system quickly and reliably.