Advanced replica management with reptor

Abstract

Data replication is one of the best known strategies to achieve high levels of availability and fault tolerance, as well as minimal access times for large, distributed user communities using a world-wide Data Grid. In certain scientific application domains the data volume can reach the order of several petabytes; in these domains data replication and access optimization play an important role in the manageability and usability of the Grid. In this paper we present the design and implementation of our replica management Grid middleware providing a high-level user and application interface to replication services. It is designed to be extensible and evolvable so that user communities can adjust its detailed behavior according to their QoS requirements. Our prototype implementation is based on the web service paradigm, in accord with the emerging Open Grid Services Architecture, and provides promising results.