An Architecture for Distributed Grid Brokering

Computational resource brokering on the Grid is the process of discovering what systems are capable of running a job, obtaining estimates for when that job may run and how much it will cost, and submitting the job to the system that best meets the users' requirements. This paper identifies how resource brokers differ from superschedulers, and describes a resource brokering architecture which is adapted to the emergent structure of the large-scale Grid. We outline the architecture of the UNICORE resource broker which is the basis of our prototype implementation, and discusses both how the existing UNICORE architecture is relevant to the wider brokering picture and what will be done in the future to bring them into closer alignment.

[1]  Ian T. Foster,et al.  The anatomy of the grid: enabling scalable virtual organizations , 2001, Proceedings First IEEE/ACM International Symposium on Cluster Computing and the Grid.

[2]  David Snelling,et al.  Explicit Trust Delegation: Security for dynamic Grids , 2004 .

[3]  Carole A. Goble,et al.  Semantic Matching of Grid Resource Descriptions , 2004, European Across Grids Conference.

[4]  Ian T. Foster,et al.  Grid information services for distributed resource sharing , 2001, Proceedings 10th IEEE International Symposium on High Performance Distributed Computing.

[5]  Werner Nutt,et al.  Relational Grid Monitoring Architecture (R-GMA) , 2003, ArXiv.

[6]  Ian T. Foster,et al.  The Anatomy of the Grid: Enabling Scalable Virtual Organizations , 2001, Int. J. High Perform. Comput. Appl..