Information Services address the challenging problems of announcing and discovering resources in Grids. Independent Grid projects have developed their own solutions to Information Services. These solutions are not interoperable with each other, target vastly different systems and address diverse set of requirements: Large-scale Grid applications require management of large amounts of relatively slowly varying metadata. E-Science Grid applications such as dynamic Grid/Web Service collections require greater support for dynamic metadata. We research Grid Information Services that support both the scalability of large amounts of slowly varying metadata and the performance demands of rapidly updated information in dynamic regions.

We propose a novel system architecture that provides unification and federation of information in Grid Information Services. The proposed system utilizes publish-subscribe paradigm and associative shared memory platform to provide an add-on architecture that interacts with existing information systems. We present an evaluation of our approach and investigate its practical usefulness. The results demonstrate that the proposed system improves the quality of information services in terms of performance and fault-tolerance with negligible processing overheads. The results also indicate that efficient decentralized Grid Information Service Architectures can be built by utilizing publish-subscribe based messaging schemes.