Uma Abordagem Autonômica para Tolerância a Falhas na Execução de Aplicações em Desktop Grids

Abstract

Computers grids are characterized by the high dynamism of its execution environment, resources and applications heterogeneity, and the requirement for high scalability. These features turn tasks such as configuration, maintenance and recovery of failed applications quite challenging and is becoming increasingly difficult to perform them only by human agents. The autonomic computing paradigm denotes computer systems capable of changing their behavior dynamically in response to changes in the execution environment. For achieving this, the software is generally organized following the MAPE-K (Monitoring, Analysis, Planning, Execution and Knowledge) model, in which managers perform the execution environment sensing activities, context analysis, planning and execution of dynamic reconfiguration actions, based on shared knowledge about the controlled system. In this work we present an autonomic mechanism based on the MAPE-K model to provide fault tolerance for applications running on computer grids, which is capable of monitoring the execution environment and, based on the evaluation of the collected data, to decide which reconfiguration actions must eventually be applied to the fault tolerance mechanism in order to keep the system in balance with the goals of minimizing the applications average completion time and to provide a high success rate in completing their tasks. This paper also describes the performance evaluation of the proposed autonomic mechanism, accomplished through the use of simulation techniques that took into account several opportunistic desktop grids typical environmental scenarios.