FAULT MANAGEMENT METHOD IN A NODE NETWORK AND PART OF THE ASSOCIATED NODE NETWORK

摘要

The invention relates to a method for managing failure in a network of nodes (2), comprising, for each considered node (2) all or part of the nodes (2) of the network performing the same calculation: first, a step local backup of the state of this considered node (21), at the level of a storage medium (31) of this considered node (21), the link (6) between this storage medium (31) and this node considered (21) being able to be redirected from this storage medium (31) to another node (23), then, a restart step: either from the considered node (21) if it is not faulty, from the local backup of the state of this considered node (21), or of an operational node (23) different from the considered node (21), if the considered node (21) is faulty, from the recovery of the backup local of the state of this considered node (21), by redirecting said link (6) between the considered node (21) and its storage medium (31) so as to connect said storage medium (31) to led it operational node (23), the local backups of these considered nodes (2), used for the restart steps, are mutually consistent so as to correspond to the same state of this calculation. Figure for the abstract: Fi gure 1