OGSA-DQP: A Service-Based Distributed Query Processor for the Grid

The Grid is an emerging infrastructure that supports the discovery, access and use of distributed computational resources. The emergence of a serviceoriented view of hardware and software resources on the grid raises the question as to how database management systems and technologies can best be deployed or adapted for use in such an environment. We argue that distributed query processing (DQP) can provide effective declarative support for service orchestration, and we describe an approach to service-based DQP (OGSA-DQP) on the Grid that supports queries over Grid Data Services (GDS)s provided by OGSA-DAI project, and over other services available on the Grid, thereby combining data access with analysis; uses the facilities of the OGSA to dynamically obtain the resources necessary for efficient evaluation of a distributed query; adapts techniques from parallel databases to provide implicit parallelism for complex data-intensive requests; and uses the emerging standard for GDSs to provide consistent access to database metadata and to interact with databases on the Grid. The service-based Distributed Query Processor is itself cast as a service referred to here as Grid Distributed Query Service (GDQS). In addition, OGSADQP employs another service for query evaluation referred to here as Grid Query Evaluation Service (GQES). As such, OGSA-DQP implements a service orchestration framework in two sense: both in terms of the way its internal architecture handles the construction and execution of distributed query plans and in terms of being able to query over data and analysis resources made available as services.