Optimizing SPARQL Query With Cache Based On Mapping Relation

With the development of semantic technology,the scale of RDF dataset is increasing year by year,which brings great challenge to the query performance of RDF data.In the face of a large number of queries and massive dataset,how to get the query result quickly and efficiently is one of the hottest issues in the current semantic web research area.The current research focuses on RDF data storage index optimization and SPARQL query optimization.In this thesis,we propose a method to use the cache.For a SPARQL query,we calculate the result set by getting the mappings between the query and the queries in the cache,and then use the cache to calculate the result set directly.For the query statement that does not exist any mappings,use the underlying storage to query the result set.In order to calculate the mappings of query statements,this thesis first proposes a method to calculate the mappings between two triple patterns,then use it to calculate the possible mapping results between two basic graph patterns.Then the mapping results are combined to get the mappings between two query statements with basic graph pattern,and then the method is given to calculate the mappings of SPARQL query with optional graph pattern and alternative graph pattern separately.Finally,the query plan is generated according to the above methods,at the same time,a cache mechanism is designed for the optimization.The experiments show that the proposed method can effectively improve the efficiency of SPARQL queries by mapping realtion and cache,and the method even more efficient when the RDF dataset is larger.