A diagnosis module based on statistic and QoS techniques for self-healing architectures supporting WS based applications

Abstract

In literature it is common to find that a self-healing architecture is made up basically of three modules: monitoring, diagnosis, and recovery. Of these three modules, the diagnosis module represents a crucial point, since in this one the state that keeps the system is established. Nevertheless, a standardized way does not exist to implement this module in this kind of architecture. In this paper we propose a strategy of implementation of diagnosis module based on statistic methods by using box plot diagrams. This technique allows us to calibrate the parameters of quality of service (QoS) in a Web services based application. This way, based on the values of QoS, the diagnosis module determines if the system is stable or if a QoS degradation is presented.