Failure Semantic of Neighbour Replication Grid Transaction Model

Real time application with replicated data in grid environment has received particular attention for providing efficient access, fault tolerance and enhance the performance of the system. This paper presents the new failure semantic to manage neighbour replication data in distributed environment. The failure semantic classes include the transaction abort and system failures. We address how to build reliable system for managing replication and transaction based on Neighbour Replication Grid (NRG) Transaction Model in terms to preserve the data consistency and support fault tolerance when exists the neighbour replica failure in quorum. In addition, we present an implementation of the propose failure semantic through NRG daemon by deploying real time application in distributed systems. The implementation shows that managing transactions on NRG through the failure semantic provides fault tolerance capabilities that allow it to withstand failure both in handling quorum locking and the transaction execution.