On Demand Replication of WSRF-based Grid Services via Cloud Computing
Germán Moltó (Universidad Politécnica de Valencia)
Vicente Hernández (Universidad Politécnica de Valencia)
With the introduction of the Globus Toolkit 4 (GT4), the service-oriented paradigm fully entered the Grid Computing arena. Since then, Grid services have turned into a suitable technology for building loosely coupled components for Grid deployments. Grid services require developers to consider fault tolerance and high availability as critical design factors to ensure successful long-term stability. This paper describes the integration of a library that allows to automatically replicate GT4-based Grid Services with a Cloud Computing back-end. Using an Infrastructure as a Service approach, the Cloud can provision GT4-based pre-configured virtual machines which are properly contextualized at boot time to deploy new Grid service replicas. This scheme allows replicated Grid Services to gain on demand scalability and increased fault tolerance.
Parallel and Distributed Computing