QUERYING RDF DATA IN THE GRID

The semantic Web provides a common framework that allows data to be shared and reused across applications, enterprises, and community boundaries. It is based on the Resource Description Framework (RDF), which is intended to provide a simple way to make statements (triples) about resources. RDF is based on the idea that the things being described (subjects) have properties (predicates), which have values (objects). Since objects can refer to the subjects of other resource statements, RDF creates relationships between resources. Large amounts of RDF data have already been created and stored in various applications and projects worldwide. For example, the UniProt Protein Database [1] contains 262 million triples, while the IngentaConnect bibliographic metadata storage [2] contains over 200 million triples. With the rapid proliferation of semantic Web technologies, there is an increasing demand for access to the large amounts of distributed RDF data. We argue that a scalable, robust access mechanism should be based on grid computing technologies. Recently, grids have applied semantic Web technologies to discover resources, manage policies, and describe provenance information, among other actions. Like the semantic Web applications, semantic grid applications usually handle a large amount of RDF data. An example is the CombeChem semantic Grid project [3], which stores more than 80 million triples in multiple databases. Therefore, providing a standard grid-based interface for accessing (the large amounts of) RDF data in the grid is important