Research On RDF Storage System Based On Cloud Computing

With the development of the Semantic Web, RDF storage system has becomea significant research topic. Traditional RDF storage system has encountered thebottleneck under the trend of the explosive growth of Internet data. Its storageand query response performance have been unable to meet the growing demand ofusers. So researchers began to focus on cloud computing which has massive datastorage and strong computing power that is considered a suitable solution to solvemassive RDF data storage problem. Research in this area is still in its infancy.The target of this dissertation is to design and implement massive RDF datastorage system. Based on the studies of early RDF data storage plan and features ofcloud computing, a suitable RDF storage plan which contains data storage modeland query strategy is proposed.Reasonable RDF data storage model plays an important role in enhancing theperformance of the management and query of RDF data. Based on the analysis ofthe advantages and disadvantages of the current RDF storage model, dissertationput forward a model of RDF data storage based on Ordered Binary Tree. The modelnot only considers the structure properties of RDF data itself but also involves thecharacteristics of data storage and processing under cloud computing environment.The model guarantees the query response efficiency of RDF data as well as reducesthe storage cost effectively. Meanwhile, the model can answer the influence ofrapid growth of RDF data easily.In the aspect of RDF data query strategy, through the research of themechanism of parallel computing framework MapReduce, combined with thecharacteristics of RDF data storage model, in order to optimize query performance,the dissertation gives the algorithm of query task generation and proves that thealgorithm can effectively improve the efficiency of query response. Based on thisalgorithm, the dissertation does further research on the characteristics of RDF datamodel based on Ordered Binary Tree and gives RDF data query scheme in detail.Compared with the traditional RDF data query scheme, this scheme makes full useof the advantages of cloud computing in the data processing, effectively improvethe query efficiency of RDF data storage system. At last, the dissertation designs the prototype system for RDF data storagemodel and query response strategy and verifies the effectiveness of the storagemodel and response strategy by experiments.