Computational Grids represent an enabling technology that permits the dynamic coupling of geographically dispersed resources (machines, networks, data storage, visualization devices, software, and scientific instruments) for performance-oriented distributed applications in science, engineering, medicine, and increasingly e-commerce. Grids provide several important benefits for users and applications: convenient interfaces to remote resources, resource coupling for resource-intensive distributed applications and remote collaboration, and resource sharing. Research in this burgeoning area embodies the confluence of high-performance parallel computing, distributed computing, and Internet computing, attracting successful research from all three disciplines. To date, application domains targeting the Grid come from science and engineering yielding strong empirical and quantitative results. In addition, NSF's TeraGrid effort, which seeks to leverage these successes to enable its Distributed Terascale Facility, will certainly encourage significant new research in Grid computing to serve the scientific communities. Many challenges must be overcome for Grids to achieve their potential.
This special issue of the Journal of Parallel and Distributed Computing (JPDC ) is dedicated to recent advances in Computational Grids that address these challenges with topics including: architectures (including P2P), fundamental building blocks, security, fault tolerance, scheduling and resource management, programming models and tools, Grid access and interface, and applications, all targeting Grid computing settings.
This special issue attracted 33 submissions. Each paper was sent to four reviewers. At least three reviews were received for every paper while some received four reviews. The papers were ranked based on the referee reports and eight papers were selected for this special issue.
The first paper in this special issue EUROGRID-European Computational Grid Testbed by Bala, Lesyng, and Erwin describe a large-scale multinational Grid testbed targeting biomedical applications. It is based on the UNICORE infrastructure, which provides a uniform access to computational resources. The EUROGRID provides seamless and secure access to site resources, with site autonomy conserved. Application specific interfaces and ''plugins'' have been developed to ease deployment on the Grid.