A new unreliable failure detector for self-healing in ubiquitous environments - Archive ouverte HAL Accéder directement au contenu
Communication Dans Un Congrès Année : 2015

A new unreliable failure detector for self-healing in ubiquitous environments

Résumé

Due to the nature of ubiquitous systems, nodes (e.g., Sensors) are frequently prone to failures. Such systems must, therefore, present self-healing capabilities in order to detect failures and make the necessary adjustments to prevent their impact on applications. In such a context, this work proposes a new and flexible unreliable failure detector, denoted as the Impact failure detector (FD), for self-healing system in ubiquitous environments. The output of the Impact FD concerns the confidence in the system as a whole. By expressing the relevance of each node by an impact factor value as well as a margin of acceptable failures of the system, the Impact FD enables the user to tune the failure detection configuration in accordance with the requirements of the application: in some scenarios, the failure of low impact or redundant nodes does not jeopardize the confidence in the system, while the crash of a high impact factor one may seriously affect it. Either a softer or stricter monitoring is thus possible. The performance evaluation results using real Planet Lab traces confirm the degree of flexible applicability of our failure detector and, that due to the margin of failure, the number of false responses may be reduced when it is compared with traditional unreliable failure detectors.
Fichier non déposé

Dates et versions

hal-01213333 , version 1 (08-10-2015)

Identifiants

Citer

Anubis Graciela de Moraes Rossetto, Carlos Rolim, Valderi Leithardt, Guilherme Borges, Claudio Geyer, et al.. A new unreliable failure detector for self-healing in ubiquitous environments. The 29th IEEE International Conference on Advanced Information Networking and Applications (AINA-2015), Mar 2015, Gwangiu, South Korea. pp.316-323, ⟨10.1109/AINA.2015.201⟩. ⟨hal-01213333⟩
142 Consultations
0 Téléchargements

Altmetric

Partager

Gmail Facebook X LinkedIn More