Building Fault-Tolerant Consistency Protocols for an Adaptive Grid Data-Sharing Service

We address the challenge of sharing large amounts of numerical data within computing grids consisting of clusters federation. We focus on the problem of handling the consistency of replicated data in an environment where the availability of storage resources dynamically changes. We propose a software architecture which decouples consistency management from fault-tolerance management. We illustrate this architecture with a case study showing how to design a consistency protocol using fault-tolerant building blocks. As a proof of concept, we describe a prototype implementation of this protocol within JuxMem , a software experimental platform for grid data sharing, and we report on a preliminary experimental evaluation