A dependability management mechanism for ubiquitous computing systems

Abstract

Dependability for a ubiquitous computing system must be guaranteed for each single component of a system and for the whole system, which - designed to fulfill a certain task - might be more than just a sum of its components. Ubiquitous computing systems must enable the testing not only each of software components separately but also of the whole system. In this paper, we propose a management mechanism for applying software rejuvenation technology into a ubiquitous computing system. It adopts the automatic monitoring scheme, the automatic analysis scheme, the autonomie plan and the execution scheme to suggest the optimal configuration alternative of a ubiquitous computing system. We validate the autonomie fault management scheme based on a workload model derived from the system log analysis. © IFIP International Federation for Information Processing 2005.