| Currently, the Semantic Web is widely used in various fields such as medicine, biology, geographic information services. However, with the advent of the era of big data and the expanding scale of big data applications, semantic data are generated at an extremely high speed. Traditional semantic data storage management techniques and systems based on relational database are unable to manage large-scale semantic data, besides that traditional semantic query optimization strategy is difficult to adapt to the large-scale semantic data query processing. In this background, using parallel computing technology to solve large-scale data storage and query problem has become a hot research issues. However, parallel computing is largely dependent on features of applications and the application itself has different complexity and diversity, which makes large-scale semantic data processing face great technical challenges.In response to these problems, on the basis of the Resource Description Framework RDF (Resource Description Framework) and RDF data query language SPARQL (Simple Protocol and RDF Query Language), the paper proposed large-scale distributed data storage and query semantic technology method based on the HBase and Redis using industry-standard OpenRDF Sesame semantic data processing framework. The method uses a hybrid index which is built on tiered storage architecture to improve semantic data query performance; On this basis, the paper further analyzes the SPARQL query processing engine, and builds a cost model to optimize query execution strategies to ensure the efficient semantic data query; In order to improve the reliability and availability of the query engine, the paper also proposes solutions for fault tolerance and scalability. Finally, the paper also designed and implemented a semantic data storage and query prototype system. Experimental results show that the large-scale data storage and query system is effective and feasible.Research work is divided into the following two parts:(1) By studying existing semantic data storage technology, the paper design a semantic data storage model, on this basis proposing a hybrid method built on the hierarchical storage mechanism and the frame, besides that the fault tolerance of the storage frame and scalable solutions are given.(2) By analyzing the process of semantic data query engine, the paper propose an optimization algorithm based on selectivity estimation with the query cost model, the paper also propose a batch query optimization strategy. |
