The Low Latency Fault Tolerance System - Computer Science > Distributed, Parallel, and Cluster ComputingReportar como inadecuado




The Low Latency Fault Tolerance System - Computer Science > Distributed, Parallel, and Cluster Computing - Descarga este documento en PDF. Documentación en PDF para descargar gratis. Disponible también para leer online.

Abstract: The Low Latency Fault Tolerance LLFT system provides fault tolerance fordistributed applications, using the leader-follower replication technique. TheLLFT system provides application-transparent replication, with strong replicaconsistency, for applications that involve multiple interacting processes orthreads. The LLFT system comprises a Low Latency Messaging Protocol, aLeader-Determined Membership Protocol, and a Virtual Determinizer Framework.The Low Latency Messaging Protocol provides reliable, totally ordered messagedelivery by employing a direct group-to-group multicast, where the messageordering is determined by the primary replica in the group. TheLeader-Determined Membership Protocol provides reconfiguration and recoverywhen a replica becomes faulty and when a replica joins or leaves a group, wherethe membership of the group is determined by the primary replica. The VirtualDeterminizer Framework captures the ordering information at the primary replicaand enforces the same ordering at the backup replicas for major sources ofnon-determinism, including multi-threading, time-related operations and socketcommunication. The LLFT system achieves low latency message delivery duringnormal operation and low latency reconfiguration and recovery when a faultoccurs.



Autor: Wenbing Zhao, P. M. Melliar-Smith, L. E. Moser

Fuente: https://arxiv.org/







Documentos relacionados