Optimisation de la gestion des points de reprise dans les réseaux mobiles

Abstract

This work, consist of study the management of checkpoints, in distributed applications executed on fixed or mobile networks. The study of check pointing protocols, dedicated for distributed applications, allow us to classify this protocols according to some facts, such as: the quantity of redo works, the number of checkpoints that we have to store, the number of hosts, implicated to participate to the recovery. Mobile computing raises many new issues, such as: lack of stable storage, low bandwidth of wireless channel, high mobility and limited battery life…These new issues make traditional check pointing algorithms unsuitable for mobile computing.