Semantic EMR Data SPARQL Query Optimization Mechanisms

As the maturing of Web2.0application and Semantic Web research, the semantic information which querys EMR in medical science and information science field development has become a hot issue. In the study, a variety of medical scholars abstract information and stores it as the Semantic Web RDF triples form, but with the development of electronic medical records semantics, exploding amount of data, which triggered a large data query efficiency.The paper, which smooths over the query effiency of semantic EHR data, designed and implemented SPARQL (Simple Protocol and RDF Query Language, SparQL Protocol and RDF Query Language) queries, which based on RDF triples of semantic EMR data and deployed on a CUDA platform. Through parallelization SPARQL query of semantic EMR data, its main function is continuous to improve query speed.The paper uses the SPARQL query strategy of parallel continuous (Paralleling Continuous-SPARQL). The policy contains three core components:a. Efficient computing with CUDA features that will SPARQL part CUDA parallel computing platform to achieve; b. To take full advantage of the hardware, in order to improve query efficiency, the SPARQL query optimization methods make it to achieve a continuous query; c. comes through Java JNI interface written in the Java language and the SPARQL query language with C libraries CUDA platform connection. Thus, through JNI in SPARQL query and create a channel between the CUDA platform, you can quickly send and receive data. In this paper, the SPARQL queries and SPARQL CUDA-based platform after comparing continuous query result:When the amount of data in a certain range, based on CUDA platform SPARQL continuous queries over the SPARQL query efficiency, and speedup the rise.